Download ROLE OF IONIZED CALCIUM AND MAGNESIUM IN CELLULOSE

ROLE OF IONIZED CALCIUM AND MAGNESIUM IN CELLULOSE
DEGRADATION BY RUMINAL BACTERIA
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree
Doctor of Philosophy in the Graduate School of The Ohio State
University
By
Maria Sol Morales Silva, D.V.M., M.S.
*****
The Ohio State University
2005
Dissertation Committee:
Dr. Burk A. Dehority, Adviser
Dr. Jeffrey Firkins
Dr. Steven C. Loerch
Dr. Mark Morrison
Dr. William P. Weiss
Approved by
____________________________
Adviser
Animal Sciences Graduate Program
ABSTRACT
To establish the ionized calcium (Ca+2) requirement for growth (G) and cellulose
degradation (CD) by rumen bacteria, cultured strains representing the principal
cellulolytic rumen bacteria were used: F. succinogenes (FS) A3c and S85, R.
albus (RA) 7 and 8, and R. flavefaciens (RF) B34b and C94. They were
incubated in complete liquid media with cellobiose for G experiments and
complete media with cellulose for CD experiments; both varied in Ca+2
concentrations. Growth was measured by OD (600 nm); and CD, by substrate
disappearance.
Results obtained with both substrates were analyzed by a
logistic function, using nonlinear regression (PROC NLIN-SAS); with cellobiose,
max growth (MG), rate of growth (RG) and lag time (LT) were calculated; and
with cellulose, extent of degradation (ED), rate of degradation (RD) and lag time
(LT) were determined. These coefficients were analyzed by ANOVA and LSM
were compared by linear and quadratic contrasts (PROC MIXED-SAS) and the
maximum of the first derivative was used to establish the cation requirement.
Different strains responded differently to the Ca+2 treatments with cellobiose
substrate.
Ca+2 concentrations had a significant effect (P<0.05) in all the
parameters evaluated, either for G (MG, RG, LT) or CD (ED and LT), with the
exception
of
RD
(P>0.05);
the
interaction
strain
within
species*Ca+2
concentrations always was significant (P<0.05) for G and CD for all parameters.
For MG, Ca+2 requirements were estimated as follow: FS-S85: 0.47mM, and RA7: >0.64 mM (tendency); whereas no estimation for A3c, B34b and C94. FS A3c
and S85 showed an absolute requirement for CD. Ca+2 requirements for ED
ii
were: FS-Ac3: >0.36; RA-7: 0.28; RF B34b and C94: >0.64 mM; whereas no
estimation for FS-S85. FS showed considerable variation in ED, perhaps due to
interactions of G and CD; therefore, NH3-free complete cellulose media was used
to separate the effect of G on CD, confirming the absolute Ca+2 requirement of
FS for CD; requirements calculated as: 0.42 and >0.64 mM for A3c and S85,
respectively. A logistic function could not be fitted for RA 8, due to the slow G
and limited CD; therefore, RA 8 was analyzed separately; by linear regression
was determined RG and RD; and a positive effect on RG (P<0.05), and no effect
on RD (P>0.05) were found related to Ca+2 concentration. The averages from
MG and ED were analyzed by ANOVA and no Ca+2 concentration effect was
found (P>0.05). Due to the response to Ca+2 from RF strains, their G was tested
for other specific cation requirements.
These showed that RF has an absolute
Mg requirement for G. Then all the strains used in the study were evaluated for
Mg requirements for G; requirements were calculated as: S85: 0.56; RA-7: 0.52;
RA-8: 0.51; RF-B34b: 0.54 and RF-C94: >0.82 mM; for FS-A3c no estimation.
For RF, CD in NH3-free media was used to evaluated the effect of Ca+2 and Mg;
thus no CD was observed with (Ca+2–free and Mg-free) and (Mg-free, plus Ca+2);
and no differences in ED were found with (plus Ca+2, plus Mg) and (Ca+2-free,
plus Mg) (P>0.05), reflecting a Mg requirement for CD. Thus, Ca+2 and Mg
requirements differ among the strains within species, as well as by the substrate
available.
iii
DEDICATION
With love and pride I dedicate this work:
In memorium of my Father
To my Family,
Don,
and friends.
iv
ACKNOWLEDGMENTS
I like to express my thanks to:
- Dr. Burk a. Dehority, my adviser, for his guideness, support and friendship.
- Drs. Jeff Firkins, Steve Loerch, Mark Morrison and Bill Weiss, for been part of
my committee.
- To Department of Animal Sciences (OSU), and it Graduate Committee, for the
support through the Research Assistantship, which permitted me do my studies.
- To Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile,
(Santiago, Chile), for their authorization and support, which permitted me came
to The Ohio State University to do my PhD program.
- To all of persons that in one or other way gave friendship, technical and moral
support in different moments through the studies and the development of my
research, all of them… my friends……
- To my family, far in Chile, for their confidence on me, for their always
unconditional support through all the challenges of the life; In memorium to my
dearest and beloved Father, who with my Mother, taught me be the values that
are my foundations.
- To Don, who suffer all the consequences from a latin woman running a PhD
program… Thanks for trust me, for your scientifical challenges and support, and
for loving me.
v
VITA
1958, May 22 ……………. Born – Santiago, Chile
1983, April ……………. D.V.M. Universidad de Chile, Chile
1987, January ……………. Academico, Dept. Fomento de la Produccion
Animal, Facultad de Ciencias Veterinarias y
Pecuarias, Universidad de Chile
1988, September …………. M.S. Universidad de Chile, Chile
PUBLICATIONS
Morales, M.S. 1983. Evaluacion nutricional mineral de cerdos criados
industrialmente en Chile. Tesis Licenciatura en Ciencias Pecuarias y Medico
Veterinarias, Universidad de Chile, Chile.
Diaz, I; M.S. Morales; A., Skoknic; R. Bravo; J.I. Egana. 1986. Evaluacion
nutricional mineral de credos criados industrialmente en Chile. Turrialba 38:299305.
Egana, J.I.; M.S. Morales. 1986. Metabolismo del nitrogeno en ruminates.
Monografias de Medicina Veterinaria 8:14-22.
Diaz, I.; M.S. Morales; J.I. Egana; A. Skoknic; P.Gecele. 1987. Concentraciones
plasmaticas y hepaticas de los principales elementos minerals en credos en
crecimiento y engorda provenientes de criaderos industrials. Avances en
Ciencias Veterinarias, 2:81-87.
Morales, M.S. 1988. Evaluacion del procesamiento de camas de broiler sobre
su valor nutritivo para rumiantes. Tesis Master Sciences
en Ciencias
Agropecuarias, mencion Produccion Animal. Universidad de Chile, Chile.
vi
Perez, P.; M.S. Morales; J.L Torres; J.I. Egana; I. Diaz. 1994. Efecto del tiempo
de ensilaje sobre las caracteristicas quimicas de la sangre bovina y de algunas
visceras y decomisos. Avances en Ciencias Veterinarias 9:16-23.
Pittet, J.; M. Maino; P. Perez, M.S. Morales. 1994. Caracterizacion del Mercado
de la carne ovina en los Estados Unidos de Norteamerica. Avances en Ciencias
Veterinarias. 9:71-72.
Egana, J.I.; M.S. Morales, M. Tonelli, J. Ojeda. 1994. Caracterizacion de la
degradabilidad ruminal de los diferentes components nitrogenados presentes en
las camas y deyecciones de aves. Archivos de Medicina Veterinaria XXVI:2934.
Morales, M.S.; J.I. Egana. 1997. Efecto del peletizado y ensilaje de las camas de
broiler sobre su valor nutritivo como laimento para rumientes. I. Evaluacion
nutricional y productive. Archivos de Zootecnia 46:1-8.
Egana, J.I.; M.S. Morales, H. Vega; M. Tonelli. 1997. Efecto del periodo de
adaptacion al consumo de camas de aves en ovinos y bovines, sobre la
degradabilidad ruminal de su contenido de materia seca y distintas fracciones
nitrogenadas. Archivos de Medicina Veterinaria XXIX :181-188.
Maino, M.; P. Perez; M.S. Morales; R. Solis; P. Melendez; A. Carmona. 1997.
Evaluacion de los sistemas comerciales de identificacion electronica en novilos
de engorda. Avances en Ciencias Veterinarias 12:93-97.
Morales, M.S.; D.L. Palmquist; W.P. Weiss. 2000. Effects of feeding whole
soybeans, dietary copper depletion, and mammary desaturation on blood and
milk triacylglycerol fatty acids in Holstein and jersey cows. Journal of Dairy
Science 83:2105-2111.
Morales, M.S.; D.L. Palmquist; W.P. Weiss. 2000. Milk fat composition of
Holstein and Jersey cows with normal or depleted copper status and fed whole
soybeans or tallow. Journal of Dairy Science 83: 2112-2119.
vii
Cabrera, R.; A. Lopez; M.S. Morales, H. Salazar; A.M. Fuentes. 2000.
Fistulacion y canulacion permanente del compartimento 1 (Rumen) en llamas
(Lama glama). Archivos de Medicina Veterinaria 32:131-138.
Hazard, S.; M.S. Morales; N. Butendieck; P. Gomez, P. Mardones. 2001.
Evaluacion de la mezcla de ensilaje de maiz con ensilaje de trebol rosado en
diferentes proporciones, en la alimentacion invernal de vacas lecheras en la
zona sur. Agricultura Tecnica 61: 306-318.
Meneses, R. P. Perez; J. Pittet; P. Galleguillos; M.S. Morales. 2001. Estrategia
de alimentacion durante la crianza de crias caprinas criollas. Agricultura Tecnica
61: 171-179.
Perez, P. ; M. Maino; M.S. Morales; A. Soto. 2001. Effect of goat milk and milk
substitutes and sex on productive parameters and carcass composition of Creole
kids. Small Ruminant Research 42: 87-94.
FIELD OF STUDY
Major field: Animal Sciences
Ruminal microbiology/Ruminant nutrition
viii
TABLE OF CONTENTS
Abstract……………………………………………………………………………ii
Dedication…………………………………………………………………………iv
Acknowledgments………………………………………………………………..v
Vita…………………………………………………………………………………vi
List of Tables……………………………………………………… ……………xi
List of Figures…………………………………………………………………...xiv
1
Introduction………………………………………….………………………….…..1
2
Literature Review…………………………………….…………………………….5
- Maximazing cellulose degradation……………………………..…………...….5
- Cellulose degradation………………………………………………….………..7
- Calcium and magnesium requirements for rumen bacteria……………..…13
- Ca and Mg concentrations in ruminal fluid/content……………………....…17
- Bacterial cell walls………………………………………………………….…..22
- Interaction between cell walls and ions………………………………….…..24
- Calcium and Bacteria………………………………………………….……....29
-Ca+2 homeostasis……………………………………………………….….30
-Calcium transport………………………………………………………..…32
-Calcium as signal………………………………………………………..…33
- Magnesium and bacteria………..………………………………………….….39
- General synthesis……………..…………………………………………….….40
3 Material and Methods…………………………………………………………...…42
- Materials…………………………………………………………………....…..42
- General Procedure…………………………………………………………….42
- Analytical Procedures…………………………………………………….…...50
- Experimental Design ………………………………………………………….52
- Statistical analysis………….………………………………………………….53
ix
4 Results and Discussion…………………………………… ……………………..60
-Ionized calcium concentrations in cellobiose and cellulose media……….60
-Growth of predominant rumen cellulolytic bacteria on cellobiose media
with different Ca+2 concentrations…………………..… ……………………65
-Cations required for bacterial growth with specific attention to
magnesium requirements… ………………………..
……………………...81
-Cellulose degradation and Ca+2 concentrations:..… ……………………...96
-Cellulose degradation with different Ca+2 concentrations in
NH3-free media……………………………………..….…………….………...109
-Cellulose degradation in NH3-free media Ca+2 and Mg-free, Ca+2-free,
Mg-free, normal divalent cation content for R .flavefaciens……………….117
-General Discussion.....………………………………………………..………119
5 Conclusions………….………………………….………………………..……….135
Appendix…....…………………………………….………………………………..137
Appendix A. Stock solutions, and different anaerobic media………………138
Appendix B. Effect of consecutive transfer on Ca+2-free, Mg-free, Ca+2-Mgfree and Normal media on growth (Max OD) by the predominant rumen
cellulolytic bacteria………………………….…………………………………..146
Appendix C. Quadratic functions used to estimate the maximum point of the
Curve……..…………….…………………………………………………………148
Bibliography…...……….…………………………………………………………150
x
LIST OF TABLES
Table 3.1 Summary for growth experiments in cellobiose and different Ca+2
concentrations………………………………………………..…….……..…………...44
Table 3.2 Summary with cellulose degradation experiments and different Ca+2
concentrations……………………………………………………..…….…………….46
Table 3.3 Summary with NH3-free cellulose degradation experiments and
different Ca+2 concentrations…………………………………………………………48
Table 4.1. Ionized Ca concentration calculated and measured by ion-electrode,
total Ca concentrations calculated and measured by atomic absorption
spectrophotometry. ……………………………………………………………………61
Table 4.2. Effect of Ca+2 (mM) concentrations on growth of Fibrobacter
succinogenes strains A3c and S85 in liquid media with cellobiose as source of
carbohydrate……………………………………………………………………………67
Table 4.3. Effect of Ca+2 (mM) concentrations on growth of Ruminococcus albus
strains 7 and 8 in liquid media with cellobiose as source of carbohydrate………70
Table 4.4. Effect of Ca+2 (mM) concentrations on growth of Ruminococcus
flavefaciens strains B34b and C94 in liquid media with cellobiose as source of
carbohydrate……………………………………………………………………………73
xi
Table 4.5.
Linear regression parameters from reciprocal of rate of growth vs.
reciprocal of Ca+2 concentrations, and Vmax and Km coefficients derivated from
the linear function……………………………………………………………………...78
Table 4.6. Growth response to divalent cations in cellobiose liquid medium by
predominant rumen cellulolytic bacteria…………………………………………….83
Table 4.7. Growth of Ruminococcus flavefaciens strains B34b and C94 with
different concentrations of Mg in liquid media………………………………………85
Table 4.8. Growth of Ruminococcus albus
strains 7 and 8 with different
concentrations of Mg in liquid media………………………………………………...88
Table 4.9. Growth of Fibrobacter succinogenes strains A3c and S85 with different
concentrations of Mg in liquid media………………………………………………...91
Table 4.10.
Linear regression parameters from reciprocal of rate of growth vs.
reciprocal of Mg concentrations, and Vmax and Km coefficients derivated from
the linear function……………………………………………………………………...94
Table 4.11. Cellulose degradation by Fibrobacter succinogenes strains A3c and
S85 incubated with different concentrations of ionized Ca+2……………………. 99
Table 4.12. Cellulose degradation by Ruminococcus albus strains 7 and 8
incubated with different concentrations of Ca+2…………………………………...102
Table 4.13. Cellulose degradation by Ruminococcus flavefaciens strains B34b
and C94 incubated with different concentrations of Ca+2………………………..105
xii
Table 4.14.
Linear regression parameters from reciprocal of rate of cellulose
degradation vs. reciprocal of Ca+2 concentrations, and Vmax and Km coefficients
derivated from the linear function…………………………………………………..108
Table 4.15. Cellulose degradation by Fibrobacter succinogenes strains A3c and
S85 incubated with different concentrations of Ca+2 and NH3-free media……..111
Table 4.16.
Linear regression parameters from reciprocal of rate of cellulose
degradation vs. reciprocal of Ca+2 concentrations, and Vmax and Km coefficients
derivated from the linear function………………….……………………………….114
Table 4.17. Extent of cellulose degradation (mg) after 60 hours incubation with
Ruminococcus flavefaciens (B34b and C94) in the presence or absence of Ca+2
and/or Mg in NH3-free cellulose liquid medium……………………………………119
Table 4. 18.
Summary of Ca+2 and Mg requirements for cellulolytic rumen
bacteria using maximum growth and extent of cellulose degradation as criterion;
Ca+2 requirements for growth and cellulose degradation, and Mg requirements
for growth………..…………………………………………………………………….132
xiii
LIST OF FIGURES
Figure 4. 1. Ca+2 concentrations in anaerobic medium measured by ion-electrode
vs. calculated (mM)…………….……………………………………………………...64
Figure 4. 2.
Ca+2 concentrations (measured with ion-electrode) vs. total Ca
concentrations (measured with AAS)……………………………………………….64
Figure 4.3. Predicted growth curves of F. succinogenes A3c with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function………………………………………………………………………..68
Figure 4.4. Predicted growth curves of F. succinogenes S85 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function……………………..………………………………………………….68
Figure 4.5. Predicted growth curves of R. albus 7 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic funtion…………………………………………………………………………71
Figure 4.6. Growth curves of R. albus 8 with different Ca+2 concentrations in
cellobiose media (Max OD vs. time, raw data)…………………………….……….71
Figure 4.7. Predicted growth curves of R. flavefaciens C94 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ……………………………………………………………………….74
xiv
Figure 4.8.- Curves of response to Ca+2 concentrations on maximum growth
(OD) by the different strains of rumen cellulolytic bacteria………………………74
Figure 4.9. Curves of response to Ca+2 concentrations on rate of growth (units
OD/hour) by the different strains of rumen cellulolytic bacteria…………………75
Figure 4.10. Predicted growth curves of R. flavefaciens B34b with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. …………………….………………………………………………....86
Figure 4.11. Predicted growth curves of R. flavefaciens C94 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ………………………..………………………………………..……86
Figure 4.12. Predicted growth curves of R. albus 7 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ………………………………………………………………………89
Figure 4.13. Predicted growth curves of R. albus 8 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ………………………………………………………………………89
Figure 4.14. Predicted growth curves of F. succinogenes A3c with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ……………………………………………………………………….92
Figure 4.15. Predicted growth curves of F. succinogenes S85 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function. ……………………………………………………………………….92
xv
Figure 4.16. Curves of response to Mg concentrations on maximum growth (OD)
by the different strains of rumen cellulolytic bacteria………….…………………..93
Figure 4.17. Curves of response to Mg concentrations on rate of growth (OD
units/hour) by the different strains of rumen cellulolytic bacteria ……….……….93
Figure 4.18. Predicted cellulose degradation curves of F. succinogenes A3c with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function…………….………………………………………………100
Figure 4.19. Predicted cellulose degradation curves of F. succinogenes A3c with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function…………….………………………………………………100
Figure 4.20. Predicted cellulose degradation curves of R. albus 7 with different
Ca+2 concentrations in cellulose media, using the parameters obtained from the
logistic function…………….………………………………………………….……...103
Figure 4.21. Cellulose degradation curves of R. albus 8 with different Ca+2
concentrations in cellulose media (raw data from replicate 1)…………………..103
Figure 4.22. Predicted cellulose degradation curves of R. flavefaciens B34b with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function…………….………………………………………………106
Figure 4.23. Predicted cellulose degradation curves of R. flavefaciens C94 with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function…………….………………………………………………106
xvi
Figure 4.24. Curves of response to Ca+2 concentrations on maximum
degradation
(mg
degraded)
by
the
different
strains
of
cellulolytic
bacteria……………………………………………………….. ……………………..107
Figure 4.25. Curves of response to Ca+2 concentrations on rate of degradation
(mg
degraded
/hour)
by
the
different
strains
of
cellulolytic
bacteria………………………………………………………………………………..107
Figure 4.26. Predicted cellulose degradation curves of F. succinogenes A3c with
different Ca+2 concentrations in NH3-free cellulose media, using the parameters
obtained from the logistic function………………………………………………….112
Figure 4.27. Predicted cellulose degradation curves of F. succinogenes S85 with
different Ca+2 concentrations in NH3-free cellulose media, using the parameters
obtained from the logistic function…………………………..……………………...112
Figure 4.28.
Maximum cellulose degradation vs. Ca+2 concentrations for F.
succinogenes strains A3c and S85, from NH3-free medium…………….………113
Figure 4.29. Rate of cellulose degradation vs. Ca+2 concentrations for F.
succinogenes strains A3c and S85, from NH3-free medium………….…………113
xvii
CHAPTER 1
INTRODUCTION
The high energy requirements of high producing dairy cows has moved
nutritionists to search for alternative energy sources and feeding systems.
Because forage alone cannot meet the higher energy requirements, concentrate
feeds are added as supplements. However, with very high producers, feeding
high levels of concentrate rich in soluble carbohydrates can lead to health and
production problems (ruminal acidosis, laminitis, low-fat milk). Thus, fats and oils
were explored as alternatives to cereals as a dense energy source. Successful
research permitted advances in this front, and today fats and oils are widely used
in diets of high producing cows. Although feeding of fats is successful, not all the
questions related with rumen function have been answered.
Unsaturated fatty acids (UFA) are toxic for the rumen bacteria and
protozoa, and ruminal biohydrogenation is the mechanism used by the bacteria
to detoxify those compounds.
As unsaturation of fatty acids increases, the
toxicity to the rumen bacteria also increases (Galbraith et al, 1971); however,
when fatty acids (FA) form salts with cations available in the rumen, toxicity is
decreased. Jenkins and Palmquist, (1982) have reported that fiber digestion was
1
negatively affected when fats/oils were included in the diets (both in vivo and in
vitro). When FA are toxic to the bacteria, they can not degrade the structural
carbohydrates (Devendra and Lewis, 1974). Fat was postulated to physically
coat the forage fiber, thus preventing bacterial attachment to the fiber (Devendra
and Lewis, 1974; Harfoot et al, 1974). Fatty acids also have the capacity to
chelate cations, such as Ca+2, decreasing the availability of this cation to the
cellulolytic bacteria that are believed to require it for growth (Devendra and
Lewis, 1974).
Numerous studies have investigated the depressed fiber
degradation in presence of fat in the diet (Palmquist and Conrad, 1981; Jenkins
and Palmquist, 1982; Ferlay and Doreau, 1995). The toxic effect of fats on
bacteria was studied by Galbraith et al (1971) and Maczulack et al (1981), and
the physical barrier effect was investigated by (Harfoot et al, 1974). Palmquist et
al (1986) explored the aspects related with Ca+2 availability in the rumen and
observed that the depression of fiber degradation was overcome when a soluble
source of calcium (CaCl2) was added to the diet, increasing Ca+2 concentration in
rumen contents. Other authors (Wagner et al, 1993), using wheat middlings in
place of corn silage, did not find increased a coefficient digestibility of the forage
and milk production when they fed diets that included CaCl2, although they were
able to find differences in Ca+2 concentration in rumen content.
Knowledge about rumen cellulolytic bacteria and the process of fiber
degradation has advanced greatly in recent years. Mechanisms by which rumen
cellulolytic bacteria attach to cellulose or structural carbohydrates are described
from genetic and molecular points of view; thus, cellulosomes, pili and their
2
structures and encoding genes are described for many rumen bacteria (Miron et
al, 2001, Bayer et al, 2004). The characterization of the structure and different
components for attaching and degrading cellulose by bacteria have shown that
cellulose binding proteins and dockerin domains have peptide sequences that
resemble the EF-hand motif protein (Michiels et al, 2002), which has the capacity
to bind cations, especially Ca+2.
This provides structural stability, heat
resistance, and in some cases increased enzymatic activity (Chavaux et al, 1990,
Chavaux et al, 1995, Bayer et al, 1998, Little et al, 2001).
Little information is available on the individual species of rumen cellulolytic
bacteria or from in vivo rumen ecosystems related to the response to Ca+2 and
what could be the requirement for this cation for fiber digestion. Many authors
(Gong and Forsberg, 1989, Roger et al, 1990, Van Soest, 1994) have suggested
that different divalent cations play unspecific roles in bacterial attachment to the
fiber, where the (+) charge of the cations interacts with the (-) charges of the
plant cell walls.
Thus, one can relate the effect of increased Ca+2 to the molecular biology,
which suggests the potential role of Ca+2 in fiber degrading systems. It is then
apparent that little quantitative information is published that can be used to
develop the idea.
Hubbert et al (1958a) observed an increased cellulose
degradation when washed rumen bacteria were exposed to increased levels of
calcium; Bryant et al (1959) worked specifically with one of the predominant
rumen cellulolytic bacteria (Fibrobacter succinogenes S85) and determined the
calcium requirement for growth with a soluble carbohydrate substrate. No studies
3
have been reported on the effect of ionized calcium (Ca+2) on growth or
cellulolytic activity, or both, which is the form of Ca required by the
microorganisms.
Presumably Ca+2 is incorporated into the cell for bacterial
structures such as the cell wall or for structures used to attach and degrade plant
fibers.
The lack of quantitative information moved me to investigate the role of
Ca+2 in cellulose digestion by the predominant rumen cellulolytic bacteria.
4
CHAPTER 2
LITERATURE REVIEW
2.1. Maximizing cellulose degradation:
Palmquist and Conrad (1982), Lopez-Guisa (1985), Palmquist et al
(1986), Wagner et al (1993), and Ferlay and Doreau (1995) have evaluated the
role of divalent cations for several different parameters related with fiber
digestion.
Palmquist and Conrad (1982) found that Ca supplementation improved
the apparent digestibility of the fiber and all the nutrients of the diet,
independently of the energy source. Lopez-Guisa (1985) evaluated the effect of
supplementing a diet based on corn crop residues and alfalfa with Cu and Co.
They observed an increased rate of both dry matter (DM) and cellulose digestion
of supplemented corn crop residues when they were incubated in situ; however
total DM degradation was not affected by the supplement.
Due to the negative effect on fiber degradation when ruminants are fed
fats or oils because their toxic effect of fatty acid on rumen microorganism
(Galbriath et al 1971), or their action as chelating agents decreasing cation
availability to the rumen microorganisms
(Devendra and Lewis, 1974), or
because of the physical coating of the vegetable fiber (Harfoot et al, 1974), many
studies were made to identify the factors that were reducing cell wall
5
degradation. Thus supplementing a 10% tallow diet with 0.96 % (3% in the
concentrate) calcium chloride to cannulated cows resulted in a lower rumen pH
at 12 hours post-feeding, a value that was highly correlated with the total VFA
concentration. Ionized calcium (Ca+2) concentration in the rumen was inversely
related to rumen pH (r = - 0.73), whereas the proportion of long chain fatty acids
(LCFA) in calcium-soap was positively related to pH (r = 0.29).
CaCl2 was
dissociated and provided higher levels of Ca+2 in rumen contents as compared
with dicalcium phosphate or limestone, and also was associated with an
increased rate of neutral detergent fiber (NDF) degradation (Palmquist et al,
1986). As mentioned, Devendra and Lewis (1974) suggested that LCFA inhibit
fiber degradation because Ca+2 availability is limited for growth of cellulolytic
bacteria. Although LCFA bound rumen Ca+2 to form insoluble soaps (Jenkins
and Palmquist, 1982), the concentration of Ca+2 in the rumen (Palmquist et al,
1986) exceeded the ruminal bacteria requirements proposed by Bryant et al
(1959).
Wagner et al (1993) tested CaCl2 (0.46% of DM) as a source of ionic
calcium in wheat middling diets (replacing corn silage) fed to dairy cows. They
found that, as Ca+2 concentration increased, ruminal pH, C2:C3 ratio, and milk fat
% all decreased; the acidic effect of CaCl2 appeared to negate any beneficial
responses of increased Ca+2 concentration. These effects were not overcome by
bicarbonate supplementation. Lack of response in this experiment was possibly
due to a minimal amount of effective fiber in the diet.
6
When fats or oils are used as an energy source in livestock feeds,
modifications occur in the rumen metabolism, among them being a decrease in
cellulose digestion. One of the aspects studied by Galbraith et al (1971) was the
beneficial effect of Ca+2, Mg+2, ergocalciferol and cholesterol concentrations on
bacteria, which appeared to reverse the negative effect of LCFA on fermentation.
Different fatty acids and different concentrations of Ca+2 and Mg+2 were
evaluated. Toxicity of the fatty acids increased with chain length and level of
unsaturation, apparently due the physicochemical properties. Both Ca+2 and Mg+2
affected lauric acid concentration, by formation of salts, but did not have the
same effect as linoleic acid. The sensitivity of bacteria to fatty acids was higher
in Gram (+) than Gram (-) cells, as well as by morphological differences, i.e.,
bacilli were more sensitive than cocci. The results suggested that supplementing
fat could influence the requirement for Ca+2 and Mg+2; thus El-Hag and Miller
(1972) had suggested increasing the supplementation of calcium when 3-5% of
fat was added to the dairy cows diet.
CaCl2 has an increased capacity to form
calcium soaps and reduce the toxic effect on rumen bacteria and improve the cell
wall digestion, when it has compared with dicalcium phosphate (Jenkins and
Palmquist, 1982).
2.2. Cellulose degradation:
The steps required for cellulose degradation by rumen bacteria were
described by Miron et al (2001), and these include: transport of bacteria to the
substrate, initial nonspecific adhesion, specific adhesion, proliferation, and
7
colonization of plant tisues.
Different mechanisms of bacterial adhesion to
cellulose/fiber have been described: cellulosome, fimbria or pili, and glycocalyx
(Miron et al 2001; Bayer et al, 2004)
In some of the cellulolytic bacteria, adhesion/attachment, and hydrolysis of
CHO involves a cellulosome structure, where an attachment system links the
bacteria and a group of enzymes with the plant cell wall (Bayer et al, 1998;
Shoham et al, 1999; Miron et al, 2001).
Cellulosome.
The
existence
of
a
“large,
discrete,
multisubunit
complex(es) which exhibits both antigenic and cellulolytic activities” was first
described in Clostridium thermocellum twenty years ago by Lamed et al (1983).
The authors proposed the concept of a “cellulosome”, a structure by which the
bacterium is attached to and degrades the insoluble substrate (Lamed et al,
1983, cited by Lamed et al, 1985). Later the cellulosome was further described,
based on biochemistry and/or genetic evidence, in many other cellulolytic
bacteria.
Among these are Clostridium cellulolyticum, C. cellulovorans, C.
acetobutylicum, C. josui, Acetivibrio cellulolyticus, Bacteroides cellulosolvens,
Ruminococcus flavefaciens, and Ruminococcus albus (Mitsumori and Minato,
1997; Ohara et al, 2000; Ding et al, 2001; Bayer et al, 2004). Also the concept
was broadened and today it includes, in addition to cellulases, enzymes that can
degrade many other cell wall polysaccharides (Bayer et al, 2004).
R. flavefaciens and R. albus also possess a glycocalyx (Miron et al 2001),
and R. albus, in addition, contains fimbria or pili (Miron et al, 2001;
Rakotoarivonina et al, 2002) as attachment mechanisms.Whereas Fibrobacter
8
succinogenes, the principal cellulolytic bacterium, does not have a cellulosome, it
has a glycocalyx (Miron et al, 2001) and presents enzymes (endoglucanases
EG2, EGF and Cl-stimulated cellobiosidase) containing a cellulose binding
protein that permits attachment to the substrate (Miron et al 2001). The presence
of a cellulosome in this bacterium is questionable; however, by transmission
electron microscopy it is possible to see protuberances around the surface of F.
succinogenes; apparently these are only blebs arising from the outer membrane.
The true mechanism for attachment of F. succinogenes remains unclear
(Chesson and Forsberg, 1997).
The cellulosome is formed by different subunits: a scaffoldin that contains
the cohesin domains, dockerins, anchoring proteins, cellulose-binding module,
and catalytic domains. Different species have a different arrangement of the
subunits, e.g., C. cellulovorans has multiple scaffoldins and C. thermocellum has
a single scaffoldin, and R. flavefaciens has a scaffoldin without identifiable CBM
and S-layer homology (Bayer et al, 2004). There are also different cohesins and
dockerins, classified according to molecular characteristics and phylogenetic
classification (Rincon et al, 2003; Bayer et al 2004).
Cohesin-dockerin interactions are species specific, first demonstrated for
C. thermocellum and C. cellulolyticum by Pages et al (1997); Ca+2 is involved in
this interaction (Choi and Ljungdahl, 1996b).
The dockerin protein contains 70 amino acid residues with 2 duplicated
segments, each with 22 amino acids. The 12 first amino acids in the duplicated
sequence show a similar feature to the Ca+2 binding loop present in EF-hand
9
motifs (Chavaux et al, 1990). EF-hand calcium-binding proteins are proteins with
two perpendicular α-helices (E and F) separated by a 12-residue loop. Residues
at 1, 3, 5, 7, 9 and 12 amino acid sequence of the loop provide the ligands for
binding Ca+2. The residues are represented by acid amino acids. EF-hand
proteins can contain between two to eight helix-loop-helix structures organized in
paired domains to facilitate their contact (Michiels et al, 2002).
Prokaryotic
proteins with EF-hand motifs are represented by extracellular polysaccharide
degrading enzymes from Clostridium, Ruminococcus and Bacteriodes (Michiels
et al, 2002).
The enzymes from Clostridium thermocellum have been described from
the molecular point of view and many of them were shown to have Ca-binding
motifs (like EF-hand motifs) related with their structural stability, permitting
formation of the cellulosome structure.
Ca-binding points are present at the
dockerin and/or catalytic domains of the cellulosome (Bayer et al 1998; Shoham
et al, 1999).
For Clostridium thermocellum the following have been described:
- cellulosome catalytic subunits: CelA, CelB, CelD, CelE, CelF, CelG, CelH, CelJ,
CelS, LicB, XynY, XynZ (Bayer et al, 1994).
- Endoglucanase CelD that contains 3 Ca-binding sites and one Zn+2 binding site,
when Ca+2 is bound to CelD it increases the substrate binding affinity, protects
against thermal denaturation (Chavaux et al, 1990), the enzyme has stable active
conformation and has greater catalytic affinity (lower Km) (Chavaux et al, 1995).
10
- Choi and Ljungdahl (1996 a, b) demonstrated that “the intact cellulosome
hydrolyzes crystalline cellulose in the presence of Ca+2 and thiols”. The
cellulosome tightly binds Ca+2 from the growth medium; CipA is involved as the
protein structure (IREs) with high affinity for Ca+2 at the scaffolding of the
cellulosome, binding several catalytic subunits at the carboxyl end of a
conserved duplicated region.
- CelS (cellobiohydrolase subunit) of the cellulosome was shown to have a
dockerin with a secondary structure that is folded into a stable tertiary structure,
mediated by binding Ca+2. This protein contains only one α-helix, comparable to
the F hand of EF-hand motif and, therefore, is not the typical EF structure (Lytle
et al, 2000). Later, Lytle et al (2001) proposed that it is possible that the dockerin
undergoes a conformational change upon binding the Ca+2 which could further
expose hydrophobic and basic residues (side-chains). This is a new idea about
cellulosome assembly.
- CBH-S8 and S11, the major cellobiohydrolase and endoglucanase,
respectively, from the cellulosome of C. thermocellum were isolated and
characterized by Bhat et al (2001).
Ruminococcus albus and R. flavefaciens also contain enzymes that have
Ca-binding sites, similar to the EF-hand group of proteins:
- Ruminococcus albus: DocVII at the dockerin domain of EGVII shows Cabinding sites (Ohara et al, 2000).
- Ruminococcus flavefaciens: EndA, XynD and XynB have dockerin-like
sequences, with 23 amino acid sequences similar to those found in cellulolytic
11
Clostridium sp. that have Ca-binding residues (Kirby et al, 1997). CesA, XynE
and XynB are esterases with Ca-binding motifs (Aurilia et al, 2000). R.
flavefaciens has a cellulosome with similar characteristics to that of C.
thermocellum, showing a Ca-binding motif at the dockerin domain (Ding et al,
2001); EndB, presenting two copies of Ca+2 binding motif similar to other
Ruminococcus domains and Clotridium sp. (Rincon et al, 2001).
Fibrobacter succinogenes, produces a number of cellulolytic enzymes and
some cellulose-binding proteins that also could have enzymatic activity against
cellulose (Chesson and Forsberg, 1997). McGavinMcGavin and Forsberg (1988)
isolated and characterized two endoglucanases in F. succinogenes S85.
Endoglucanase 1 (EG-1) was the most active enzyme, representing 32% of the
total endoglucanase activity. The addition of CaCl2 (1-10 mM) or MgCl2 (1-10mM)
increased its activity.
Endoglucanase 2 (EG-2) represented 11% of total
endoglucanase activity, and no variation in its activity was observed when Ca+2
or Mg+2 was increased (McGavin and Forsberg, 1988). EG-2 exhibited specific
binding to acid-swollen cellulose, whereas neither EG-1 nor EG-2 bound
crystalline cellulose (McGavin and Forsberg, 1988; McGavin et al, 1990). EG-2
was later renamed as endoglucanase F (EG-F), because the Cel F gene
encodes this protein (Malburg et al, 1997). EG-F belongs to glycosyl hydrolase
family 51 and has unique endoglucanase activity (Malburg et al, 1997; Mitsumori
and Minato, 2000). The N-terminal cellulose-binding domain of EG-F has two
repeated regions that have similar amino acid sequences arranged in tandem;
the first region binds cellulose (Mitsumori and Minato, 2000). Mitsumori et al
12
(2002) demonstrated, using recombinant proteins of EG-F, that these repeated
regions are involved in the binding capacity of the enzyme, and Ca+2 improved its
binding to carboxymethylcellulose. However, no calcium-binding motif was found
in the amino acid sequence of the recombinant cellulose-binding protein
(designated as AD2).
2.3. Calcium and Magnesium requirements for rumen bacteria:
Hungate (1966) considered that the mineral allowances for ruminants
estimated from Blaxter (1952) should supply largely the Ca and Mg in vitro
growth requirements of B. succinogenes in pure cultures. However, in vitro
requirements may be lower than in vivo requirements, where much more
complex substrates need to be metabolized. Also, he considered that rumen
bacteria may not tolerate large unbalances of single minerals; although the
typical mineral concentrations in the rumen content may be higher than the
bacteria requirements, these requirements may be higher than those reported for
single strains due the higher amount of substrate needing to be metabolized.
Bryant et al (1959), using Bacteroides succinogenes strain S85 (now
Fibrobacter succinogenes), evaluated the mineral requirements for growth, and
demonstrated the need for PO4-3, NH4+, Mg, Ca, K and Na in the medium. When
PO4-3, K, Mg, NH4+ or Ca were deleted singly, little or no growth occurred. For
Ca and Mg, they reported a requirement for maximum growth of 0.05 mg and
0.01 mg per 5ml of medium, respectively. This converts to a requirement of 0.25
mM of total Ca and 0.082 mM of total Mg.
13
Caldwell et al (1973) evaluated the inorganic growth requirements for
Bacteroides amylophilus and found that this microorganism required for growth:
Na, PO4, K and small amounts of Mg, while no requirement for Ca, Co, Mn, Cl, or
SO4 could be demonstrated. In further studies, Caldwell and Hudson (1974)
evaluated the sodium growth requirement for predominant rumen bacteria using
pure cultures and found an obligate requirement for growth (both yield and
growth rate) for all the species and strains tested. The deletion of Na from the
media could not be replaced by Rb, Li or Cs.
Later, Caldwell and Arcand (1974) evaluated the mineral requirements for
many species from the genus Bacteroides. When Ca was deleted from the
media, growth of B. fragilis, B. oralis 7CM and B. fundiliformis 12290 was
reduced. When they added only Ca (28 μM) or Mg (10 μM) to cation-depleted
media, normal growth was not observed, but when Ca and Mg were added
together, normal yield was achieved by B. ruminicola subsp. ruminicola B18, B.
fragilis subsp. vulgatus 8483 and B. succinogenes S85. The addition of Co or Mn
to a medium with Ca and Mg reduced the growth of B. succinogenes S85,
suggesting a complex interaction among the different cations in this strain.
Mineral requirements for cellulose degradation were first investigated in
vitro by Hubbert et al (1958a, 1958b). In the first study (1958a), the authors
evaluated the addition of Ca, Mg, S, Mn, Fe, Cu, Co, Zn and B to in vitro
systems, and in the second (1958b) they evaluated the effects of K, Na, Rb, Li,
and Cs.
In both studies the authors used washed suspensions of rumen
microorganisms. Under these conditions they found that K, but not Na, was
14
necessary for cellulose digestion (Hubbert et al, 1958b). Among the minerals
evaluated in the first study, Ca, Mg, and S were recognized as nutrients that
must be added in artificial media. They suggested as optimal concentrations: 50300 μg/ml of CaCl2, 20-160 μg/ml of MgSO4 and 10-500 μg/ml of Na2SO4. Higher
concentrations 450, 320 and 1000 μg/ml, respectively, were considered to be
toxic because cellulose degradation was decreased (Hubbert et al, 1958a). By
calculation, the optimal range for Ca was 0.34 to 2.04 mM; and the toxic
level,11.23 mM.
Later, Barth and Hansard (1962), using washed ruminal microorganisms
as inoculum, confirmed that Ca is required for cellulose degradation in vitro;
when phytin was the source of P, Ca had a detrimental effect on cellulose
degradation, but not when P was from an inorganic source.
The authors
concluded that the ratio of Ca:P must be considered for supplementation.
Whether in vitro or in vivo, they found the ratio must not be higher than 4:1.
Martin et al (1964) reported a depression in cellulose degradation when
Mg was not supplemented to the ration (in vivo) or to the medium (in vitro).
Uesaka et al (1967), cited by Durand and Kawashima (1980) reported a
positive effect on cellulose degradation when the concentration of CaCl2 was
increased from 10 to 100 μg /ml of medium.
Bales et al (1978) evaluated the degradation of DM from sorghum stalks
incubated in vitro with different levels of Ca, P, Mg and urea, using washed
rumen bacteria. As Ca or Mg concentrations increased (10 to 190 ppm, and 50
to 150 ppm, respectively) DM digestion was decreased (P<0.05), suggesting a
15
toxic effect from high Ca or Mg concentrations on DM digestion; considering that
the principal components of the stalks are the structural carbohydrates, this
decreased DM digestion can be interpreted as a decreased fiber digestion.
Gong and Forsberg (1989), studying the behavior of the principal
cellulolytic bacteria, determined that Fibrobacter succinogenes requires either Ca
or Mg for adhesion to cellulose. Conversely, Roger et al (1990) found in their
experiments that F. succinogenes does not require Ca+2 for adhesion, but
Ruminococcus flavefaciens requires Ca and Mg for the cellulose adhesion
process.
In 1973, Bryant remarked on the little research done about mineral
requirements for cellulolytic bacteria, that this may be because of an assumption
that they have similar requirements to other bacteria. However, he pointed out
that, “contrary to most non-marine bacteria so far studied, B. succinogenes has a
requirement for Na and relatively high need for Ca”.
He suggested that mixed
cultures must need K, Na, Ca and phosphate for optimal cellulose digestion, and
maybe Mo, Fe and Zn are also required. Subsequently, in their review, Durand
and Kawashima (1980) commented: “It is not certain whether all bacteria have a
Ca requirement for growth. Ca deficiency can cause growth defects and the
formation of pleomorphic forms, and might alter metabolic processes requiring
Ca+2 activated exoenzymes as α-amylase.
It can be considered that
microorganisms require a smaller amount of Ca+2 than Mg+2.”
16
Very little additional research on mineral requirements of bacteria
occurred after this time, which is based on the idea that all microorganisms have
the same requirements, as reflected in the review of Durand and Kawashima
(1980).
The research about mineral requirements by rumen bacteria and their
different functions clearly has had a lack of direction; although certainly many
individual efforts have been made, only isolated information has been obtained.
2.4. Ca and Mg concentrations in ruminal fluid/content:
Poutiainen (1970) measured Ca and Mg concentrations and their changes
in the rumen fluid of cattle. Ca and Mg were determined by atomic absorption
spectrometry. The author found significantly different concentrations among
sampling times and also sampling location in rumen (P<0.05). The lower
concentrations for Ca and Mg were obtained at 0 and 12 hours after feeding, and
the peak was obtained at 3-4 hours after feeding. The Ca concentrations varied
from 2 to 3.5 mM and for Mg from 2 to 4 mM (4 to 8 meq/l).
Lower
concentrations were found at the bottom of the rumen, probably due to dilution.
Nel and Moir (1974) evaluated the effect of supplying Ca and/or P by
ruminal or duodenal infusion in sheep, with the idea to identify the site of
absorption of these in the gastrointestinal tract. As part of the methodology they
determined the total Ca concentration of centrifuged rumen liquor by atomic
absorption spectrophotometry. Samples from control and non-ruminally infused
Ca animals ranged from 0.46 to 0.68 mM of Ca (mean 0.60 mM). Concentration
of Ca in the rumen was 1.04 mM when CaCO3 was infused ruminally.
17
Playne et al (1978) evaluated the disappearance of different minerals from
hays from Medicago sativa, Stylosanthes humilis, Chloris barbata and
Heteropogon contortus using the nylon bag technique at 5 incubation times (12,
24, 48, 72 and 168 h). The minerals that remained in higher proportions after 168
h of incubation were P, Ca and N, (40% of DM) whereas Mg showed an
intermediary resistance (20% of DM), and K had the least (~5% of DM). The
minerals from Medicago sativa were the most soluble, whereas Heteropogon
were the most insoluble. The most rapid release from the forage occurred during
the first 24 h of incubation in rumen. Later relative concentrations increased
because their release was slower than the loss of dry matter.
Bennink et al (1978) followed the postprandial changes in mineral
composition of the rumen fluid of cattle and sheep fed with different diets. Ca
and Mg concentrations were affected by their intake. As can be expected, Ca
concentrations were increased 3 to 9 times when alfalfa was fed. The Ca and Mg
concentrations at 24 h were related to intake. Negative correlations were
observed for Ca and Mg concentrations and pH (increased solubility with lower
pH). The ranges of Ca and Mg concentrations preprandially were: 0.4–1.7 and
1.0–4.4 mM, respectively. The range of means from all the samples from 24
hours from the 6 diets was: 0.39–10.0 and 1.79–10.10 mM, for Ca and Mg,
respectively. Concentration peaks varied among diets, ranging from 1 to 3 hours
postprandial. These values represent the total soluble minerals, because Ca and
Mg were determined in the rumen fluid filtered through cheesecloth and
determined by atomic absorption spectrometry.
18
Mackie and Therion (1984) reported a summary of studies that measured
Ca and Mg concentrations in rumen; no information about methods used to
measure the cations nor about conditions of sampling was provided. The ranges
for Ca and Mg were 0.75 – 12 mM and 0.40 – 6.5 mM, respectively.
Rumen content of soluble Ca, determined by atomic absorption
spectrometry, from steers fed 88% concentrate and varying levels and sources of
calcium, no Ca, CaCO3 (0.4%), CaCl2 (0.4%), CaCO3 (2.5%) were: 55; 165; 64;
and 239 mg Ca/l, respectively (Goetsch and Owens, 1985).
Because the composition of plant species influences mineral availability in
the rumen, Emanuele and Staples (1990) evaluated the minerals released from 6
different forage species using in situ technique. Immediate and maximal releases
in rumen were 29 and 70 % for Ca, and 82 and 95% for Mg, respectively. The
highest Ca concentration (1.46 % DM) was in alfalfa and 46.7% of this was
released immediately and 12.6% was released slowly, requiring 12 to 18 hrs to
reach the maximal release. Ca concentrations in other species were lower, and
Ca was released over longer time. Dwarf elephant grass bound ruminal Ca,
increasing the Ca concentration by ~20 %. The measurements for Ca and Mg
forage content were made using atomic absorption spectrometry. Grace et al
(1977) contributed to the understanding of the dynamics of Ca, Mg, P and K in
rumen digesta and the factors that affect it.
The authors used eight sheep
cannulated in both rumen and duodenum/ileum and fed Lolium perenne to
evaluate the extent of the association of Mg, Ca, P, and K with organic
19
constituents of the digesta in relation to different sites of the gastrointestinal tract
and their apparent absorption. Rumen content was subdivided into different
pools: water-soluble, diluted-alkali-insoluble, and dilute-alkali-soluble fractions.
Ca, Mg and P were 48.0, 57.4, 45.9 % water-soluble in the diet and 54.6,
56 and 50 % as water-soluble compounds in the rumen, respectively. The watersoluble fraction contained the ionic and bound forms of Ca, Mg, and P. The
dilute-alkali-insoluble fraction contained Ca+2, Mg+2 and P: 37.9, 17.2 and 24%,
respectively.
The alkali-insoluble fraction from rumen digesta contained N,
pectin, hemicellulose, cellulose, 72% H2SO4-insoluble and ash (Grace et al,
1977). Among those compounds pectin and lignin have the ability to bind Ca+2
and Mg+2(Molloy and Richards, 1971;, Somers 1973; Torre et al 1992).
When they evaluated the effect of changes in pH on binding capacity of
the different minerals with the dilute-alkali-soluble and dilute-alkali-insoluble
fractions, they found an increased Mg binding to the alkali-insoluble fraction
when pH increased; as the pH decreased, more Mg and Ca became associated
with the water-soluble fraction. Similar results occurred when the medium was
diluted; at higher dilution, higher content of Ca and Mg was in the soluble
fraction. When Ca+2 was increased in the medium, it had a negative effect on
Mg+2 binding to the alkali-insoluble fraction, probably by competition at binding
sites (Grace et al, 1977).
The results from Grace et al (1977) confirmed the earlier reported effect of
pH on cell wall binding ions formulated by Storry (1961), but also clarify the issue
related to the components of the water-soluble fraction, both ionic and bound
20
forms, and why total Ca of this fraction does not reflect the amount of ionized Ca,
which is the interchangeable form and the one that is required by the
microorganisms.
Ruminal volatile fatty acids (VFA) can bind calcium, forming Ca salts that
are water soluble compounds, but not ionized (Windholz, 1976). These probably
readily exchange with the ionized fraction as Ca+2 becomes bound to fiber or
long chain fatty acids, thus tending to maintain an equilibrium of the ionized form.
Ionized Ca concentrations in rumen fluid were monitored by Palmquist et
al (1986) and Wagner et al (1993), who both measured Ca+2 with ion-electrode.
They demonstrated that rumen Ca+2 concentrations are affected by the diet,
source of Ca, and time after feeding, and the range varied widely between 0.25
mM to 7 mM.
Palmquist et al (1986) reported for cows supplemented with tallow and
either CaCl2 or Ca2PO4 + limestone, an average from different dietary treatments
of 0.6 mM at 0 h and 1.5 mM of Ca+2 at 10 h after feeding, when concentration
reached a plateau. The mean rumen content of ionized calcium when CaCl2 or
dicalcium phosphate + limestone was supplemented were: 1.82 and 0.62 mM,
respectively.
Wagner et al (1993) observed, in dairy cows, the peak concentration of
Ca+2 ocurred at 6 hours after feeding, independently of the type of diet evaluated,
declining to basal levels at 12 hours after feeding. When high forage diets were
supplemented with CaCl2 or CaCl2+ NaHCO3 the values for Ca+2 at 3, 6 and 12
hours after feeding were: 3.75, ~7, and 0.75 mM, respectively. When wheat
21
middlings with CaCl2 were fed the Ca+2 concentrations were: ~7, ~7 and 0.75
mM; whereas wheat middlings + CaCl2 + NaHCO3 were supplemented the
concentrations were: 1.5, 3.75 and 0.25 mM. For wheat middlings+NaHCO3, the
values were: 1.5, 2.5 and 0.25 mM of Ca+2.
Ferlay and Doreau (1995), feeding cows supplemented with rapeseed oil,
evaluated the ultrafiltrable calcium concentrations from rumen fluid by AAS;
observing a decreased concentration of calcium when fats/oil were fed to the
cows; two diurnal peaks were observed related with time of feeding rather than
the diet.
Concentrations reported were: 77.6, 21.1, 25.7 mg/l ultrafiltrable
calcium for control, and the two diets supplemented with oil, respectively.
2.5. Bacterial cell walls:
The cell walls of Gram (+) and Gram (-) bacteria differ in structure and
composition. The Gram (+) bacteria have a cytoplasmic membrane and a thick
peptidoglycan (also called mureine) layer. In contrast, Gram (-) bacteria have
more complex cell walls: the cytoplasmic membrane, periplasmatic space,
peptidoglycan, and outer membrane (bi-layer), which contains an inner layer of
phospholipids, lipoproteins and proteins and an outer layer of lipopolysaccharide
(40%) and proteins (60%) (http://www.ugr.es/~eianez/Microbiologia/05pared.htm
3/15/05). The cell walls of archea are the most simple, containing only the
cytoplasmic membrane and S-layer (Mayer, 1999).
22
The peptidoglycan layer also contains a matrix of teichoic acid, teichuronic
acid and lipoteichoic acid. The latter is present in all the Gram (+) bacteria,
whereas the former is present in only some. Teichuronic acid is produced by
certain Gram (+) bacteria when they are phosphate-restricted, and this acid
contains anionic residues that are able to bind divalent cations, particularly Mg+2
(http://www.ugr.es/~eianez/Microbiologia/05pared.htm 3/15/05).
The polyglycerol phosphate-teichoic acid, which is located in the
periplasmic region of the cell, also can bind divalent cations and, together with
teichoic, and teichuronic acids, form a cation-exchange system in the cell wall
(Ghuysen and Shockman, 1973).
Bioynthesis of peptidoglycan occurs in 3 stages: “Peptidoglycan
precursors made on a uridylic acid cytoplasmic carrier (stage 1) are transferred
from uridylic acid to an undecaprenyl phosphate membrane carrier (stage 2) and
to a final acceptor in the expanding wall peptidoglycan (stage 3). At some point
during the later stages of this process, the nascent peptidoglycan undergoes
cross-linking, which is required to make it insoluble, and covalently linked
accessory wall polymers are attached.” At the first stage the synthesis of the
nucleotide precursors is catalyzed by enzymes that require ATP and either Mg+2
or Mn+2. At the second stage the translocation of N-acetyl-muramyl-pentapeptide
requires Mg+2 and is stimulated by K+ and other monovalent cations (Ghuysen
and Shockman, 1973).
23
Other structures are S-layers, and when present, they are the most
external layer of the cell wall; they function as protective coats, for bacterial
adhesion, as surface recognition, as ion traps, and as scaffolding for enzymes
and virulence factors. They are ubiquitous, present in Gram (+) and (-) bacteria
(Sleytr and Beveridge, 1999, Mayer, 1999).
S-layers are self assembled
protein/glycoprotein subunits, occurring in a monolayer, planar lattice (Sleytr and
Beveridge, 1999,); Ca+2 is required for the subunit attachment in Caulobacter
crescentus, (Walker et al, 1994 cited by Sleytr and Beveridge, 1999), and
Campylobacter fetus subspecies fetus (Dworkin et al, 1995 cited by Sleytr and
Beveridge, 1999).
2.6. Interaction between cell wall and ions.
Among the factors that affect the initial nonspecific adhesion of bacteria to
substrate are ionic or hydrophobic associations that involve van der Waals
forces, Ca+2, Mg+2, or the double ion layer of K that can neutralize the repellent
ionic negative charge between cell walls and bacteria (Van Soest, 1994, Miron et
al, 2001); also temperature, pH, and age of the bacteria can affect this interaction
(Miron et al, 2001).
Molloy and Richards (1971) reported the Ca+2 and Mg+2 binding capacities
of different polymers from the cell wall of Holcus lanatus. They found that pectin
and lignin had higher affinity for both cations, while hemicellulose and cellulose
24
had very low capacities to bind these cations. Increasing pH from 6.3 to 7.24
increased the Ca+2 binding percentage from 28 to 31.7 % of pectin. Lignin bound
20.7% of Ca+2 at pH 6.2. Pectin and lignin bound 3.8 and 12.5 % of Mg+2,
respectively.
Calcium ions have a particular affinity for plant cell walls, reacting with the
carboxyl groups of the pectic components of cell walls; the adhesion is lost when
tissues are washed with acid solutions (Somers, 1973).
Torre et al (1992) evaluated in vitro the Ca+2 binding capacity of different
components of the plant cell wall under different conditions of varying pH, Ca+2
concentration, pectin, lignin and cellulose concentration. Binding by lignin
appeared to be the strongest, due the proton-ionizable functional groups of lignin;
binding was affected by pH (3.5-7.5, binding increased when pH increased),
lignin and Ca+2 concentrations. Apparently, Ca+2 binds lignin according to the
multiple equilibrium theory, with formation of a stoichiometric bond. The affinity
of Ca+2 with cellulose or pectin was very weak, and may occur only as Ca+2
adsorption on the polysaccharide surface. The pH did not affect Ca+2 binding by
cellulose and pectin, suggesting that the binding capacity of cellulose is not
associated with the hydroxyl residues of its polymer chains.
By using low-
methoxylated pectin, it was shown that pectin does not bind Ca+2 through the
carboxyl groups of the D-galacturonic acid. Based on these studies, cellulose
does not bind Ca+2, and lignin is the strongest compound in vegetable cell walls
25
that binds Ca+2.
With regard to pectin, the information is less conclusive,
perhaps because of differences in chemical composition of pectin used in
different studies or different methodology to establish binding.
Research from Van Soest’s laboratory developed techniques to measure
the cation exchange capacity (CEC) of feedstuffs (Allen et al, 1985). Among
different resources evaluated, Whatman No 42 paper and solka floc showed the
lowest CEC. A broad classification of feeds, from low to high CEC, was listed:
corn silage, straws, brans, grass hays and legume hays. Among feedstuffs with
intermediary NDF content, almond hulls and rapeseed showed the highest CEC
values (Allen et al, 1985).
McBurney et al (1986) reported a significant
correlation between CEC and lignin content (0.693 to 0.843), lignin:ADF ratio
(0.531 to 0.742) and nitrogen content (0.5 to 0.579), using the Pr III method
under two pH: 3.5 and 7.0, respectively. CEC of feedstuffs increased when pH
increased, in this form, the CEC from dietary fiber influences the rumen
environment by buffering pH and affects the osmotic pressure because of the
capacity to bind ions.
The affinity between bacterial cell walls and different ions also has been
studied. Fitt et al (1974) reported that isolated rumen bacterial cell walls can take
up Ca+2 and Mg+2; the binding increased with increased cation concentrations,
and the presence of Ca+2 reduced the binding of Mg+2, and vice-versa. Fitt and
Hutton (1974) reported that K+ also interfered with Mg+2 binding.
26
Beveridge and Murray (1980) reported that phosphodiester groups of
teichoic acid in Bacillus subtilis bind Mg+2 strongly (carboxyl groups from teichoic
acid are the principal binding site for different ions); from the total mineral content
of the native cell walls (21.108 umol of metal/mg) Ca+2 represented 1.9%, while
Mg+2 was 39%, Fe+3, 17% and Cu+2, 14.2%.
Later, Beveridge et al (1982), studying B. licheniformis, found that the
principal compounds in the cell wall able to bind metals were teichoic and
teichuronic acids. On native cells walls Ca+2 was 53%; Mg+2, 24%; and Na+, 11%
of the total mineral content (0.208 umol of metal/mg).
Although both bacteria studied were Gram (+) and from the same genus,
they had different cell wall composition with regard to the type, amount and
structural disposition, thus affecting the capacity to bind different metals.
More evidence about the differences in cell wall composition among Gram
(+) bacteria is the variability in response to Gram staining (Gram variable).
Beveridge (1990) reported that the main reason for variability in Gram staining is
aging of the bacteria, differences in composition (bacteria with higher percentage
of peptidoglycan stain Gram (+)), thickness of the cell wall, and the presence of
S-layers, as occur in Bacillus, Butyrivibrio and Clostridium.
When Beveridge et al (1982) evaluated the capacity to bind metals from
the cell envelope of E. coli, they found different affinities among the 32 metals
tested: Ca+2, Cu, Na, K, Fe+2 were bound in small amounts (<0.1umol/mg DM);
Mg+2, Zn, Fe+3, Mn, were bound in intermediary amounts (0.1 to 0.4 umol/mg
DM); and in large amounts were bound hafnium and osmium. The binding site
27
corresponds principally with the “polar head group regions of the constituent
membranes or long peptidoglycan layer”. In Gram (-) bacteria the outer
membrane contains Mg+2 and sometimes Ca+2 as integral compounds
(Beveridge, 1981; cited by Beveridge et al, 1982) and they do not bind as much
metal as Gram (+) bacteria do (Beveridge et al, 1982).
Durand and Kawashima (1980), considering the differences in cell
wall/envelope composition between Gram (-) and Gram (+), indicated that the
former must have lower requirements for Ca+2 and Mg+2 than do Gram (+)
bacteria.
Well known is that pH affects the binding capacity of the walls, both plant
and bacterial. Storry (1961) reported that the maximum adsorption of calcium to
the cell wall occurs at neutral pH, but it is decreased as the pH decreases. This
was supported by studies from Durand and Kawashima (1980) and McBurney et
al (1986).
This point is important to consider, because when pH conditions are
maintained stable in a well buffered medium, normally close to neutral pH, the
CEC is maximum and the minerals are bound to the cell wall and are not
available to interchange with the media or with other bacteria. Under rumen
conditions, pH decreases after feeding, principally because of increasing
concentration of volatile fatty acids.
This pH condition permits part of the
minerals from the feed and bacteria to became ionized and available for the
microflora and microfauna (Palmquist et al, 1986). Also the ions can form new
28
salts by interacting with other nutrients or metabolites (calcium soaps, VFA-salts,
etc) (Sukhija and Palmquist, 1990) that can be soluble or insoluble in the
ultrafiltrable rumen content (Grace et al, 1977).
2. 7. Calcium and Bacteria
The role of Ca+2 in bacteria has been relegated to the cell wall and
external environment of the bacterial cell, principally for activating external
enzymes (Silver, 1977, Smith, 1995; Norris et al 1996). Ca+2 is required for
growth of photosynthetic bacteria and for Azotobacter; but routinely Ca+2 is added
to the culture media of all bacteria without knowing whether or not it was needed
for growth (Silver, 1977). The active participation of Ca+2 in cell wall synthesis is
more related with sporulation and encystment processes (i.e. A. vinelandii
require 0.4mM for optimum encystment), rather than with the synthesis of the cell
wall (Silver, 1977); Ca+2 can bind chemical compounds at the cell wall.
Lately, diverse roles have been identified for Ca+2 in bacteria; it has been
shown to participate in chemotaxis, thermostability, motility, activation of
intracellular enzymes and cell cycle regulation processes (Ordal, 1977;
Rampersaud et al, 1991; Tisa and Adler, 1995; Yu and Margolin, 1997; Herbaud
et al, 1998; Sudom et al, 2003).
29
2.7.1. Ca+2 homeostasis:
Intracellular Ca+2 concentrations are maintained low in bacteria as in
eukaryotes, due to the presence of specific Ca+2 channels (Silver and Kralovic,
1969 cited by Gangola and Rosen 1987), which permit bacterial cells to export
excess calcium (Deves and Brodie, 1981).
Intracellular concentrations of Ca+2 in bacteria were measured first in E.
coli by Rosen and Gangola (1987), who used a specific dye, fura-2
pentaacetoxymethyl ester, preceded by treatment with EDTA to facilitate
permeation. The free-Ca+2 concentrations were similar to those found in
eukaryotic cells (90 ± 9 nM of Ca+2 with extracellular range of 10uM – 10 mM).
Total
Ca
concentrations
increased
with
increasing
extracellular
Ca+2
concentrations; however intracellular free-Ca+2 was independent of the
extracellular concentration, which they proposed could be due to a large pool of
bound Ca and an efficient system to extrude Ca+2 from the cell. When the energy
status of the cells was normal, the intracellular concentration of Ca+2 did not vary
with extracellular Ca+2 concentrations (10 uM to 10mM). When cells were energy
depleted and extracellular Ca+2 concentrations increased, the intracellular
content of Ca+2 in the cells increased until it became equilibrated with the
extracellular concentrations, demonstrating energy dependence to extrude the
Ca+2 by Ca+2/H+ or Ca+2HPO-4/H+ antiporters.
Jones at al (1999) measured the intracellular free Ca+2 in E. coli by using
the bioluminescent protein aqueporin and demonstrated that internal Ca+2
increases very rapidly in response to increased extracellular Ca+2 concentrations.
30
Depending on the growth status when they were exposed to the increased Ca+2,
internal Ca+2 concentration was normalized rapidly or slowly, without mediation of
protein
synthesis.
When
the
bacteria
were
adapted
to
higher
Ca+2
concentrations, they were able to increase the intracellular Ca+2 faster. Using
different channel blockers it was found that apparently the bacteria have two
Ca+2 influx systems, one that is affected by La+3 and the other that is not affected.
The La+3 sensitive channel corresponds with the voltage-activated poly-βhydroxybutyrate/phosphate Ca+2 channels described by Reusch et al (1995, cited
by Jones et al, 1999).
Research from Campbell’s laboratory (Jones et al, 1999) shows that the
periplasmic space of E. coli (Gram-) can act as a “buffer” for Ca+2 concentrations
between the intracellular and the extracellular conditions. The highly anionic
periplasmic membrane-derived oligosacharides are responsable for this Ca+2
store; when these are modified the intracellular free-Ca+2 concentration is
predictable, because it is generated by a Donnan potential. They also showed
that the periplasmic space can store three to sixfold the Ca+2 concentration
detected in the external medium when this is in micromolar concentrations. At
millimolar concentration in the external medium, the periplasmic space can store
Ca+2 similar to concentrations in the external medium (Jones et al, 2002).
31
2.7.2. Calcium transport:
Channels are intramembrane proteins of the cytoplasmic membrane; also
at the outer membrane there are proteins that can act as channels, called porins
(Mayer, 1999). Deves and Brodie (1981) in their review on active transport of
Ca+2 in bacteria, describe three mechanisms for Ca transport in bacteria: Ca+2/H+
antiporter (E. coli, A. vinelandii, M. phlei), Ca+2/Na+ antiporter (H. halobium) and
ATP (S. faecalis).
The presence of specific Ca channels, temperature-dependent in E. coli,
was first described for bacteria by Silver and Kralovic1969 cited by Gangola and
Rosen, 1987). Later, other channels were described for Ca+2 efflux: a Ca+2/H+
antiporter (Tsujibo and Rosen, 1983; cited by Gangola and Rosen, 1987) and a
Ca+2HPO-4/H+ antiporter (Ambudkar, Zlotnick and Rosen, 1984, cited by Rosen
and Gangola, 1987).
More recent information shows that polyphosphate / poly-(R) -3hydroxybutyrate are homopolymers, ubiquitous in eukaryotic and prokaryotic
cells; these are present at the bacterial plasma membranes, concentrations
varying according to physiological stage and environmental conditions. They
form ion channels in planar lipid bilayers that are highly selective for Ca+2 over
Na+ at physiological pH. Apparently these complexes can “make” pores at the
membrane, thus facilitating ion transport. These also can collaborate to regulate
specific cation transfer systems where other molecules are involved (Reusch,
2000). All these mechanisms are used to maintain low intracellular Ca+2
concentrations.
32
Ionophores: “An ionophore is a lipid-soluble small molecule that functions
to transport an ion across otherwise impermeable membranes” (Silver 1977).
One of the better known ionophores is monensin; the mechanism of action was
described by Russell and Strobel (1987). Monenesin is an antiporter that
facilitates
the
movement
of
monovalent
cations
across
membranes:
K+(eflux)/H+(influx) and Na+(influx)/H+(efflux).
There are a variety of other compounds that can act as ionophores for
divalent cations; also, specific Ca+2 ionophores are available: X537A (Lasalocid,
Hoffman-La Roche), A23187 (Eli Lilly and Company), and beauvericin (Silver,
1977). The affinity for Ca+2 or Mg+2 or other divalent cations differ among
ionophores, for X537A: Ba+2>Sr+2> Ca+2>Mg+2 (Pressman, 1973; cited by Silver
1977), whereas for A23187: Mg+2> Ca+2 (Reed and Lardy, 1972; cited by Silver
1977). Tetronasin is an ionophore that acts as an antiporter to interchange
divalent cations with H+, permitting the influx of two protons and the efflux of one
Ca+2 or Mg+2 (Newbold et al, 1988).
2.7.3. Calcium as a signal:
Numerous papers show a signaling role for Ca+2 in prokaryotes in the
same fashion as this cation does in eukaryotic cells. The capacity of prokaryotes
to regulate intracellular Ca+2 concentrations (Rosen and Gangola, 1987; Tisa and
Adler, 1995, Jones et al, 1999, Jones et al, 2002) supports this idea. Ca+2 in
prokaryotes has a role in cell division, chemotaxis, thermostability, cofactorallosteric factor in enzymes etc, as will be described below.
33
In a review on the role of Ca+2 in bacteria, Norris et al (1996) concluded
that intracellular Ca+2 concentrations and protein phosphorylation must be part of
the structural change mediators in prokaryotes, just as they are in eukaryotes.
Further, intracellular levels of Ca+2 increase when E. coli is in cell division
process and during adaptation to different environment; and Ca+2 was suggested
to participate in the regulation of the cell cycle via protein modifications that
control the enzoskeleton, thus acting as a metabolic switch. Enzoskeleton was
defined by Norris et al (1996) as “the ensemble of proteins that interact with one
another and with membranes and nucleic acids to form extended estructures
within the cell”.
These ideas are supported by research from Yu and Margolin (1997), who
found that Ca+2 regulates GTP binding and FtsZ hydrolysis. FtsZ, a tubulin-like
GTPase, forms a ring at the cell mid point before cytogenesis begins. FtsZ is
essential to generate a cell division point and forms part of the enzoskeleton. It
has the capacity to polymerize and depolymerize by adding or extracting Ca+2
from the in vitro media. E. coli cells without FtsZ can grow without dividing,
because no formation of the division septum, resulting in long multinucleated
filaments.
L-forms of E. coli lack a normal peptidoglycan sacculus (wall-less); despite
this, they can grow and divide. E. coli L-form NC-7 was used to demonstrate that
the enzoskeleton is controlled by Ca+2 (Onoda et al, 2000). When the medium
lacked Ca+2, the cells stopped division, become spherical, swelled, formed
34
vacuoles, and lysed. When Ca+2 was reestablished in the medium (1.2 mM), cell
growth was restarted (50 uM was not sufficient to restore growth). The L-form
cells had FtsZ in lower concentrations than in the wild type (Onoda et al, 2000).
A role for intracellular Ca+2 to activate intracellular enzymes was reported
by Sudom et al (2003). They found under in vitro conditions in E. coli that
phosphoenolpyruvate carboxykinase (PepCK), a key regulatory enzyme for
gluconeogenesis, requires ATP, Mg+2 and Ca+2 as cofactors. Ca+2 bound the
enzyme at the active site in a different location than does Mg+2; when trypsin was
added the Ca+2 activation of PEPCK was abolished. It is thought that PEPCK
activation by Ca+2 may play a role during glycogen synthesis at the stationaryphase of E. coli (Sudom et al, 2003).
Concerning a chemotactic role of Ca+2, Ordal (1977), using Bacillus
subtilis, demonstrated that changes in the intracellular Ca+2 concentration affect
the direction of flagellar rotation; high concentrations cause the bacteria to
tumble, whereas at lower concentrations, they swim. Later, Tisa and Adler
(1995), using wild-type and chemo-receptor mutants of Escherichia coli, tested
the changes in intracellular Ca+2, using the dye fura-2 as indicator. When
attractans were included in the media, the intracellular Ca+2 concentration
increased briefly and the bacteria tumbled until Ca+2 recovered its steady state
condition. When repellents were added, intracellular Ca+2 decreased briefly, and
bacteria swam until Ca+2 concentrations were reestablished to normal levels.
35
When neutral substances were added, no response was observed. When
mutants with no chemoreceptors were tested, no response to the different
chemo-stimulants/inhibitors was observed.
The tumbling and the swimming behaviour is regulated by the expression
of CheY, a cell wall protein, in the phosphorylated or unphosphorylated state,
respectively, responding to high or low levels of intracellular Ca+2 (Lukat et al
1990 cited by Tisa and Adler (1995).
OmpC and OmpF are outer membrane porin proteins that change in
response to changes in the osmolarity of the external environment. The
transcriptional regulation of these proteins depends on EnvZ, a protein of the
inner membrane with a periplasmic receptor domain and cytoplasmic signaling
domain. EnvZ signals to OmpR and it regulates the OmpC and OmpF expression
(Forst and Inouye, 1988; cited by Rampersaud et al, 1991). It was demonstrated
that Ca+2 stimulates EnvZ phosphorylation, followed by transfer of phosphate
from EnvZ to OmpR, which activates the porin genes (Rampersaud et al 1991).
Van Asselt and Dijkstra (1999) reported the presence in E. coli of a lytic
transglycosylase, Slt35, which has a motif similar to the EF-hand motif; this was
able to bind Ca+2, thus enhancing the thermostability of the enzyme.
Slt35
catalyses the cleavage of β(1,4)-glycosidic bonds in peptidoglycan at the
bacterial cell wall, allowing the insertion of new units into the growing cell wall,
creating pores for proteins and transport, and acting as a cell wall zipper during
cell division.
36
Outer membrane lipases, such as phospholipase A in E. coli and
phospholipase C in B. cereus and C. welchii, are stimulated or inhibited by Ca+2,
depending on specificity of the respective phospholipase (Silver, 1977), thus
affecting the permeability or fluidity of the membrane. Snijder and Dijkstra (2000)
reported that outer membrane phopholipase A is encoded by the pldA gene,
which is widespread among Gram (-) bacteria and associated with pathogenicity.
Among rumen bacteria, B. fibrisolvens outer membrane phospholipases do not
appear to be sensitive to Ca+2 stimuli; rather, they are sensitive to the presence
of cysteine, dithiothreitol and mercaptoethanol. No information about any other
rumen Gram (-) bacteria was found (Hazlewood and Dawson, 1975 and 1976;
cited by Harfoot, 1981).
Other Ca+2 binding proteins in B. subtilis were reported by Herbaud et al
(1998): flagellin, hydroperoxido reductase (AhpC), ribosomal protein (B-L9) and
two major cold shock proteins (CspB), but in none of these was found EF-hand
motifs in their amino acid sequences.
In eukaryotes, EF-hand proteins are ubiquitous and calmodulin is the
representative proteins of this group; their regulatory function is well known for
many different enzymes and ion channels (Michiels et al, 2002). Calerythrin was
the first protein with EF-hand motifs found in bacteria (Leadly et al, 1984), from a
Gram (+) bacterium Saccharopolyspora erythrea; later, calsymin from Rhizobium
etli, Gram (-) bacterium was identified (Xi et al, 2000).
37
Michiels et al (2002) searched several databases and identified many EFhand calcium-binding proteins in prokaryotes; among them a group of sixteen
proteins with at least two EF-hand motifs were found, but they do not have a
clear function, with the exception of calerythrin, calsymin (mentioned earlier), and
Asp24 (an acid shock protein) from Brucella abortus, which play an infectious
role. A second group of prokaryotic proteins with EF-hand motifs was
represented
by
extracellular
polysaccharide-degrading
enzymes
from
Clostridium, Ruminococcus and Bacteriodes; a characteristic of this group of
proteins is that only the second α–helix loop is present. A third group includes
the periplasmic D-galactose-binding proteins that participate in glucose and
galactose transport.
Earlier, the presence of EF-hand proteins belonging to this third group
were described: periplasmic glucose/galactose-binding proteins from E. coli
(Vyas et al, 1987 cited by Van Asselt and Dijkstra, 1999) and S. typhimurium
(Zou et al, 1993 cited by Van Asselt and Dijkstra, 1999) membrane-bound lytic
transglycosylase B (MltB) from Gram (-) bacteria (Van Asselt and Dijkstra, 1999).
Ca+2 depletion reduced 2 fold the sugar affinity of D-Galactose receptor in E. coli,
due an allosteric structural change in the protein (Luck and Falke, 1991).
38
2. 8. Magnesium and bacteria.
Magnesium is the most abundant intracellular divalent cation and its
essential role in all living cells is broadly recognized (Smith and Maguire, 1998).
In prokaryotes, Mg has many different cell functions: bacterial chemotaxis,
enzyme cofactor, maintain integrity of the cell wall/cellular membrane and growth
(Jasper and Silver, 1977; Smith and Maguire, 1998). The role may differ among
bacterial species (Jasper and Silver, 1977).
Part of the Mg is present at the cell wall (Eagon et al, 1965; cited by
Jasper and Silver, 1977, Durand and Kawashima, 1980), but most is bound to
the ribosomes and is required to maintain their structure (Tissieres et al, 1959;
cited by Jasper and Silver, 1977). The ribosomes of E. coli are highly sensitive to
Mg starvation, and, when starvation occurs, bacterial growth stops because
protein synthesis and DNA synthesis decrease.
E. coli and many Gram (+)
bacteria show morphological changes after short periods of Mg starvation,
becoming filamentous. A. aerogenes is less sensitive than E. coli and can resist
Mg starvation of 20 hr or longer, maintaining DNA synthesis. Salt tolerance of S.
aureus is reduced when intracellular concentrations of Mg is decreased (Jasper
and Silver, 1977).
The intracellular concentration of Mg appear to be similar between Gram
(-) and Gram (+) bacteria; they have an absolute growth requirement for Mg+2
that cannot be replaced by other ions (Webb, 1968 cited by Jasper and Silver,
1977).
39
Jasper and Silver (1977) reported that Mg+2 is accumulated selectively by
active transport across membranes; however, transport and homeostasis of Mg
in prokaryotes remains poorly understand (Moncrief and Maguire, 1999). Three
different transport systems are described for Salmonella typhimurium and E. coli:
CorA, MgtA and MgtB and one regulatory system PhoP/PhoQ, which respond to
extracellular Mg+2 concentrations (Moncrief and Maguire, 1999; Chamnongpol
and Groisman, 2002).
2.9. General Synthesis:
As mentioned in the introduction, the complex interaction among different
nutrients (fatty acids, Ca+2, structural carbohydrates) and bacteria determine a
variable animal response. No conclusive information can explain why fiber
digestion is higher when fats/oils are supplemented with a source of ionized
calcium. The degradation of insoluble carbohydrate from cell walls is a subject of
fundamental
importance
in
ruminants,
and
amazing
advances
in
the
understanding of the process had been made; molecular biology without doubt
had contributed to identifying the complex structures that participate in the
attachment and fiber degradation by the different species of rumen cellulolytic
bacteria; but the lack of information in other aspects related with the rumen
bacteria function is evident.
There is enough evidence concerning the
participation of Ca+2 in the different structures for cellulose attachment and its
degradation by rumen cellulolytic bacteria.
However, there is not available
information that can demonstrate a specific requirement for function of rumen
40
cellulolytic bacteria, other than the information generated by Bryant et al (1959)
for B. succinogenes S85 (actually F. succinogenes). The same is true for
requirements for a numerous other mineral nutrients for bacterial growth. The
results from in vitro and vivo experiments suggest a positive effect on cellulose or
fiber degradation resulting from rumen cellulolytic bacteria increased calcium
concentration in the media, but the experiments were made with a pool of rumen
microorganisms, no species identification, and the calcium concentrations used
were highly variable among experiments, and no ionized calcium was determined
(Ca+2).
Thus this study was undertaken with the purpose to explore the following
hypotheses:
- Fibrobacter succinogenes, Ruminococcus albus and R. flavefaciens,
predominant rumen cellulolytic bacteria, require ionized Ca for growth or
cellulose degradation or both.
- Fibrobacter succinogenes, Ruminococcus albus and R. flavefaciens,
predominant rumen cellulolytic bacteria, differ regarding their requirements
of ionized Ca for growth or cellulose degradation or both.
41
CHAPTER 3
MATERIALS AND METHODS
3.1. Materials
3.1.1. Bacterial Species and Strains utilized:
Fibrobacter succinogenes, strains S85 (Bryant and Doetsch, 1954) and A3c
(Dehority, 1963)
Ruminococcus albus, strains 7 (Bryant et al, 1958) and 8 (Hungate and Stack,
1982).
Ruminococcus flavefaciens, strains B34b (Dehority, 1963) and C94 (Bryant et al,
1958).
3.1.2. Media (Composition of all media utilized in the different experiments are
described in Appendix A).
-Rumen-glucose-cellobiose-agar (RGCA) slants (Bryant and Burkey, 1953)
-Cellobiose complete liquid medium (Scott and Dehority, 1965)
-Cellulose complete liquid medium (modified from Scott and Dehority, 1965)
-Phosphate buffer
Media with varying concentrations of Ca+2, Mg and other divalent cations were all
prepared as modifications of the cellobiose and cellulose complete media.
3.2. General Procedure
Medium preparation, transfer of bacteria, inoculum preparation and
inoculations were performed under a stream of O2-free CO2 (Hungate, 1950
modified by Dehority 1969).
42
Stock cultures: All cultures were obtained from RGCA slants stored at -60°
C. Transfers were made to new slants, which were incubated at 39° C and
transferred several times until the culture showed very active growth. In general,
at the time that any strain was being used, it was transferred every day to a new
slant. Those strains not in use were kept refrigerated at 5° C and transferred
every 3 to 4 days.
Optical density (OD) was measured at 600 nm with a Spectronic® 20.
3.2.1. Protocol to determine ionized calcium requirements for growth in
cellobiose liquid medium.
The different ionized Ca concentrations used were: 0, 0.02, 0.04, 0.08,
0.16, 0.32, and 0.64 mM for F. succinogenes and R. albus, whereas for R.
flavefaciens 0, 0.02, 0.06, 0.09, 0.1 and 0.64 mM were used. Normal medium
routinely used in the laboratory served as control (Scott and Dehority, 1965,
Appendix A).
To prepare the pre-inoculum a complete cellobiose low Ca+2 liquid
medium (0.04 mM Ca+2) was inoculated with a loop of bacteria from a RGCA
slant. The culture was incubated overnight at 39° C, and the next morning 1 ml
was transferred into a complete Ca+2-free cellobiose liquid medium and
incubated until it reached an OD between 0.6 and 0.7. Enough of this culture
was added to cellobiose complete liquid Ca+2-free medium to attain an OD of 0.1.
Each experimental tube was then inoculated with 0.1 ml of the culture.
Experimental tubes and non-inoculated tubes (blanks) were incubated at 39º C
and after the lag period, the growth was monitored by OD at varying intervals.
The amount of 0.1 M CaCl2 (Orion®) used to meet the different levels
desired in the culture media were 0, 0.47, 0.93, 1.87, 3.73, 7.46 and 14.92 ml per
liter of medium to obtain 0, 0.02, 0.04, 0.08, 0.16, 0.32 and 0.64 mM Ca+2.
Table 3.1 lists the organisms, Ca+2 concentrations, replicates and range of
time over which the experiments were monitored.
43
Species
F. succinogenes
F. succinogenes
R. albus
R. albus
R. flavefaciens
R. flavefaciens
Strain
Range
of
monitoring time
(hours)
Ca+2 mM
A3c
0, 0.02, 0.04,
0.08, 0.16, 0.32, Duplicated, 1
0.64 and Control replicate.
(0.36)
0 to 24, 0 to 39
S85
0, 0.02, 0.04,
0.08, 0.16, 0.32,
0.64 and Control
(0.36)
Duplicated, 1
replicate.
0 to 22, 0 to 36
7
0, 0.02, 0.04,
0.08, 0.16, 0.32,
0.64 and Control
(0.36)
Duplicated, 1
replicate.
0 to 15, 0 to 13
8
0, 0.02, 0.04,
0.08, 0.16, 0.32,
0.64 and Control
(0.36)
Duplicated, 1
replicate.
0 to 215, 0 to 219
B34b
0, 0.04, 0.06,
0.08, 0.09, 0.1, Duplicated, 1
0.64 and Control replicate.
(0.36)
0 to 24, 0 to 22
C94
0, 0.04, 0.06,
0.08, 0.09, 0.1, Duplicated, 1
0.64 and Control replicate.
(0.36)
0 to 47, 0 to 49
Table 3.1. Summary for growth experiments in cellobiose and different Ca+2
concentrations.
44
3.2.2. Protocol to determine ionized calcium requirements for cellulose
degradation.
The different ionized Ca concentrations used were: 0, 0.04, 0.06, 0.08,
and 0.12 for F. succinogenes and 0, 0.02, 0.04, 0.08, 0.16, 0.32, and 0.64 for R.
albus and R. flavefaciens; The normal cellulose liquid medium routinely used in
the laboratory served as control (Scott and Dehority, 1965).
Pre-inoculum, inoculum and inoculation of tubes were the same as
previously described for Ca+2 growth experiments.
Experimental tubes were incubated at 39º C. Five samples were taken at
different times, i.e., 0 hour and 4 later times depending on extent of visible
cellulose digestion (Table 3.2). F. succinogenes and R. flavefaciens strains were
run in duplicate tubes (2 racks, each one containing a set of culture tubes with
the various Ca+2 concentrations, control and blanks), and the experiments were
replicated 2 times. R. albus 7 and 8, were replicated 1 times. Blanks culture
tubes, not-inoculated, were included and incubated.
Table 3.2 lists the organisms, Ca+2 concentrations, replicates and
sampling times for cellulose digestion.
45
F. succinogenes
A3c
0, 0.04, 0.06, 0.12, Duplicated, 2
and Control (0.36)
replicates
Sampling
times (hours)
0, 45, 55, 65,
80
F. succinogenes
S85
0, 0.04, 0.06, 0.12, Duplicated, 2
and Control (0.36)
replicates
0, 21, 30, 40,
58
7
0, 0.02, 0.04, 0.08,
Duplicated, 1
0.16, 0.32, 0.64 and
replicates
Control (0.36)
0, 17, 27, 39,
54
8
0, 0.02, 0.04, 0.08,
Duplicated, 1
0.16, 0.32, 0.64 and
replicates
Control (0.36)
0, 60, 132,
207, 275
R. flavefaciens
B34b
0, 0.02, 0.04, 0.08,
Duplicated, 2
0.16, 0.32, 0.64 and
replicates
Control (0.36)
0, 21, 30, 40,
70, 142
R. flavefaciens
C94
0, 0.02, 0.04, 0.08,
Duplicated, 2
0.16, 0.32, 0.64 and
replicates
Control (0.36)
0, 30, 42, 66,
121
Species
R. albus
R. albus
Strain
Ca+2 mM
Table 3.2. Summary with cellulose degradation experiments and different Ca+2
concentrations
46
3.2.3. Protocol to determine ionized calcium requirements for cellulose
degradation in NH3-free medium.
To eliminate the confunding growth effect on cellulose degradation a
cellulose NH3-free media was prepeared. Only F. succinogenes, strains S85 and
A3c were used in this experiment, because these strains were the only one
showing certain response to Ca+2 concentrations in normal NH3 media. The
ionized Ca concentrations used were: 0, 0.02, 0.04, 0.08, 0.16, 0.32, and 0.64
mM.
The procedure to prepare the inoculum was similar to that for the Ca+2
cellulose degradation experiments, but at least 8 culture tubes were grown in the
low Ca+2 liquid medium because a larger inoculum was used. The inoculum
tubes were centrifuged at 900xg for 15 to 20 min, and the supernatant was
decanted and discarded.
Phosphate buffer was added to re-suspend the
bacterial cells, the culture was mixed, OD determined and enough buffer was
added to reach 0.7- 0.75 OD. One ml was used to inoculate each culture tube of
NH3-free cellulose media.
Experimental tubes were incubated at 39º C. A total of four samples were
taken, at 0 hour and 3 later times depending on visual cellulose disappearance
(Table 3.3). Each sample was taken in duplicate (2 racks, each containing a
complete set of culture tubes with the different treatments, control and blanks).
The experiments were replicated 1 time. Blanks (i.e., un-inoculated culture tubes)
were included and incubated.
Table 3.3. lists the organisms, Ca+2 concentrations, replicates and
sampling times.
47
Ca+2 (mM)
Species
Strain
F. succinogenes
A3c
0, 0.02, 0.04,
In duplicate, 1
0.08, 0.16, 0.32,
replicate
0.64
S85
0, 0.02, 0.04,
In duplicate, 2
0.08, 0.16, 0.32,
replicate
0.64
F. succinogenes
Sampling
times (hours)
R1: 0, 12, 24,
48
R2: 0, 18, 24,
48
R1: 0, 12, 24,
48
R2: 0, 18, 24,
48
R1= Replicate 1, R2=Replicate 2
Table 3.3.
different Ca
Summary with NH3-free cellulose degradation experiments and
+2
concentrations
3.2.4. Protocol to determine divalent-cation requirements for growth in
cellobiose liquid medium.
F. succinogenes, strains S85 and A3c, R. albus,
strain 7 , and R.
flavefaciens, strain B34b were used in this experiment.
Using cellobiose as substrate, the treatments were complete media
(Control), divalent-cation-free (CaCl2, FeSO4, MgSO4, MnSO4, and ZnSO4 were
deleted), Ca+2 and Mg free, Ca+2-free, Mg-free, Fe-free, Mn-free, Zn-free, plus
Ca+2, plus Fe, plus Mg, plus Mn, plus Zn, plus Ca+2 and Mg. Concentrations for
each cation added were the same as those used by Scott and Dehority (1965).
The procedure was similar to other growth experiments. The inoculum
preparation varied because the pre-inoculum was started with a loop full from a
slant to a cellobiose liquid divalent-cation deficient medium (1/8 the concentration
of normal medium), incubated overnight at 39° C and the next morning 1 ml was
48
transferred into complete cellobiose medium that was free of divalent cations and
incubated until an OD between 0.6 to 0.7 was reached, depending on the strain
involved.
The pre-inoculum (0.6 to 0.9 ml) was added into cellobiose complete liquid
divalent-cation-free medium to attain 0.1 OD and from this 0.1 ml was used to
inoculate each culture tube (containing 7 ml of liquid medium).
Experimental tubes were incubated at 39° C, and after the lag period the
growth was monitored by optical density at 600 nm (Spectronic® 20). Readings
were taken every 1 or 2 hours, depending on growth of the bacteria, until they
reached maximum growth. Blank tubes (non-inoculated) were incubated
simultaneously. Each experiment was run in duplicated and replicated 1 time.
3.2.5. Protocol to determine magnesium requirements for growth in
cellobiose liquid medium.
The Mg concentrations evaluated were: 0, 0.05, 0.10, 0.20, and 0.82 mM.
Normal medium routinely used in the laboratory served as Control (0.41 mM)
(Scott and Dehority, 1965). Mg was supplied from MgSO4.
Inoculum preparation was similar to the other experiments; however, in
this experiment, cellobiose Mg-deficient liquid medium (0.05 mM Mg) was used
to prepare the preinoculum and cellobiose Mg-free liquid medium to prepare the
inoculum.
Incubation and monitoring of growth was similar to the other growth
experiments. Each experiment was run in duplicated and replicated 1 time.
49
3.2.6.
Protocol
to
determine
ionized
calcium
and
or
magnesium
requirements for cellulose degradation, NH3-free cellulose medium.
R. flavefaciens, strains B34b and C94, were used in this experiment, with
Ca+2-free, Mg-free, and Ca+2-Mg-free NH3-free cellulose liquid medium. An NH3free cellulose liquid medium containing all divalent cations served as control.
The procedure to prepare the pre-inoculum was similar to the cellulose
degradation experiment with F. succinogenes in NH3-free medium, except Ca+2,
or Mg or Ca+2 and Mg were the cations omitted.
Experimental tubes were incubated at 39º Cº. A total of four samples
were taken, at 0 hour and 3 later times depending on visual cellulose
disappearance.
Each sample was taken in duplicate (2 racks, each one
containing a complete set of culture tubes with the different treatments, control
and blanks). The experiment was replicated 1 time. Blanks, culture tubes noninoculated were included and incubated.
3.3. Analytical procedures:
Cellulose residue:
Following the procedure of Hiltner and Dehority (1983), the culture tubes
with cellulose were centrifuged at ca. 900xg for 10 minutes, supernatant removed
by suction, the residue was re-suspended in 5 ml of acid detergent fiber solution,
tubes capped with aluminium caps, heated at 100° C in the autoclave for 60
minutes, the residue was washed twice with boiling deionized water; the
suspension was transferred to a tared test tube, centrifuged, supernatant was
removed by suction, and the tube was dried in an oven at 100° C until the weight
was stable.
Weight of tube plus residue - weight of empty tube = cellulose residue.
cellulose residue at different sampling time (mg) - Cellulose residue of blank or 0
time (mg) - = mg degraded at each sampling time.
50
Determination of calcium concentrations:
Ionized Ca in the medium was supplied from CaCl2 (0.1 M Calcium,
Ionplus-Orion®), and different Ca+2 concentrations in cellobiose and cellulose
media were determined with an ion-meter (Accumet® AR25), using a specific
calcium electrode (Thermo Orion®, Model 93-20) with a single junction reference
electrode (Thermo Orion®, Model 90-01). To 30 ml of media, 0.6 ml of 4M KCl
as an ionic strength adjustor (ISA) was added to determine Ca+2 concentrations.
Blanks and standards were treated in a similar manner.
Both cellobiose and cellulose media were analyzed to determine total
calcium concentrations. The standard calcium solution and all samples were
treated with 1N HCl; cellulose samples were lightly centrifuged to sediment the
cellulose and an aliquot of supernatant was used for analysis; Standards and
samples had LaCl3 added to avoid interference from other minerals. Total Ca
concentrations were determined by Atomic Absorption Spectrophotometer
(Varian®) following standard analytical conditions for measuring calcium as
established by the equipment manufacter.
Determination of magnesium concentrations:
Total content of Mg concentrations in cellobiose liquid media, used for Mgrequirement experiments were determined using an Atomic Absorption
Spectrophotometer (Varian®); Samples and standard were treated with 1N HCl,
similar to calcium samples; also, LaCl3 was added to avoid interference with
other minerals. Results confirmed the calculated concentrations (data not
shown).
Glassware:
All the glassware used in the experiments was washed as normal for the
laboratory. Subsequently it was soaked with 20% HCl for 24 hours and rinsed 3
times with deionized water. The objective of this treatment was to remove all
possible mineral contamination.
51
3.4. Experimental design:
For growth and cellulose degradation Ca+2 requirements, an incomplete
factorial design was used, 2(3) x 6/8; factors were 2 strains whithin species (3),
and 6 to 8 Ca+2 levels.
The 2 strains of each species were run simultaneously, but each species
was run at a different time; then the replications were considered as Block 1 and
Block 2. Tube 1 and Tube 2 were the experimental units in growth experiments,
because the OD was measured on the tube unit. Tube 1 and tube 2 were the
experimental units for cellulose degradation, because at each sampling time one
tube was taken from each of the duplicate racks.
For cellulose degradation in NH3-free medium, a factorial 2x8 design was
used, the two strains (A3c and S85) of F. succinogenes, were assigned to 8
different Ca+2 concentrations.
As in the other experiments, replications were
considered as Block 1 and Block 2. Tube 1 and Tube 2 were the experimental
units, because the cellulose degradation was measured in each.
For Mg growth requirements, a complete factorial design was used,
2(3)x6; factors were: strains (2) within species (3) and Mg concentrations (6).
The 2 strains of each species were run simultaneously, but each species
was run at different times; then the replications were considered as Block 1 and
Block 2. Tube 1 and Tube 2 were the experimental units, because the OD was
measured on each tube.
For divalent-cation absolute requirements, an incomplete factorial design
was used, 2(3)xn, where factors were strains (2) within species (3), but only 1 for
R. albus), and n treatments. The experiment was run in duplicate and repeated
twice. The experimental unit was the culture tube, because OD was measured
for each.
The experimental design for ionized calcium and or magnesium absolute
requirements for cellulose degradation in NH3-free cellulose media for R.
flavefaciens was a factorial design 2x4, 2 strains and 4 treatments.
experiment was run in duplicate and replicated 1 time.
52
The
3.5. Statistical analysis:
As general statement, statistical significances will be accepted at P≤0.05;
and a tendency will be accepted when 0.05 < P ≤ 0.15.
Growth experiments, response to different Ca+2 and Mg concentrations:
Due to the form of the growth curve, a non-linear regression analysis was
used; the data were evaluated using the model from Zwietering et al (1990) that
describe a sigmoid curve similar to bacterial growth. The model is
OD =
A
1+ exp
4*B / A * (C - time) + 2
where, A, B and C parameters, respectively, represent:
A: the maximum growth, in this case maximum OD (units OD)
B: the rate of growth, (units OD/h)
C: lag time (hours)
using NLIN PROC from SAS. Due to the different number of observations and
the different times when the OD were registered in each experiment, and the
requirement from the analytical program, it was not possible use NLIN MIXED for
the analysis.
All the species and strains evaluated, with the exception of R. albus 8 for
Ca+2 growth, fitted the Zwietering model (P<0.05).
Parameters A, B and C, and their respective standard errors for each tube
were obtained; the parameters were weighted by the reciprocal of the standard
error; these were analyzed by variance analysis (ANOVA) using PROC MIXED
(SAS). Fixed effects were: Ca+2 concentrations, species, strain within species;
the interactions: strain within species * Ca+2 concentrations; random effects were:
block within species and tube within block. The extended and reduced models
were compared; because these were different (P<0.05), the extended model was
53
used. The homogeneity of the variance was evaluated, and for Ca+2 growth
response A, B, C, parameters were analyzed with an unstructured variance test.
Least squares means (LSM) were obtained and compared with linear and
quadratic contrasts ( PROC MIXED, SAS).
To identify the requirement of Ca+2 or Mg for growth, was used the
maximum growth, rate of growth and lag time as criteria. When linear effect was
obtained as statistically significant, no estimation of requirement can be made,
because end point was not attained. When quadratic effect was obtained, non
linear analysis was made to find the break point and the plateau of the function
(PROC NLIN, SAS). When this method did not solve, maximum of the first
derivative of the quadratic function (PROC REG, SAS) was used to estimate the
requirements.
For Mg growth response an extended model with unstructured variance
test was used for A and B, and a reduced model for C.
Data for RA-8 growth curve did not fit the logistic model, and in this case,
the data were not included in the overall analysis. It was analyzed with linear
regression to determine the rate of growth (PROC REG, SAS); then the “b”
coefficients or slopes obtained were evaluated by ANOVA.
If significant
differences were found, the least means differences test was applied (GLM,
SAS) to identify effects of Ca+2 concentrations on rate of growth.
Maximum
growth obtained from raw data, were analyzed by ANOVA, and no differences
were found (P>0.05).
Concepts related to enzyme function and substrate affinity were applied to
the data, using the Lineweaver-Burk method of plotting enzyme kinetics (Price
and Dwek, 1984). Assuming that the bacteria correspond to the “enzyme” and
Ca+2 the “substrate”, and rate of growth to the “velocity of the reaction”, this
approach appears be useful as a tool to study the bacterial affinity for Ca+2 under
the experimental conditions of the study. The affinity constant (Ks) corresponds
to the amount of substrate necessary to attain ½ of the μmax (maximum rate of
growth). Using the reciprocal of the least squares means of rate of growth (1/B)
54
of each treatment (as Y) plotted vs. the reciprocal of Ca+2 concentrations
(1/[Ca+2]) (as X), a linear function was estimated (PROC REG, SAS), thus
1/μmax (when X = 0) and -1/Ks (when Y = 0) were identified to obtain the μmax
and Ks for each strain. Ranking the Ks values, it is possible to establish the
tendency in the affinity of different strains for Ca+2.
Extended Statistical model used to study the effect of Ca+2 or Mg on
bacterial growth on cellobiose liquid media:
Y = [Ca+2] + species + strain (species)+ block + tube(block) + [Ca+2]*strain
(species) + error
Y ijklm = Ci + Spj + St(sp)k:j+ Bl:k + T(B)m:l + C*Spij + C*St(sp)i k:j+ εijklm
Where: C ([Ca+2]),
i = 1, 2, 3, 4, 5, 6, 7, 8 or
C ([Mg]),
i = 1, 2, 3, 4, 5, 6
Sp (Species),
j = 1, 2, 3
St(sp) (Strain (species)), k = 1, 2
B (Block),
l = 1, 2
T(B) (Tube(Block)),
m = 1, 2
C*Sp, interaction Ca+2 with species
C*St(sp), interaction Ca+2 with Strain within Species
ε (Error),
i,j,k,l,m ( 0, σ2e )
Cellulose degradation experiments:
For cellulose degradation, where different tubes were sampled in different
chronological times, the duplicate rack was the experimental unit. Two racks
were incubated in each replicate. Each rack contained tubes to establish the
basal cellulose content at time 0 for each treatment, and 4 tubes for each Ca+2
concentration at each of the 4 sampling times. Control tubes, uninoculated tubes
were incubated simultaneously.
55
The profile of the data from cellulose degradation was similar to a growth
curve. Because the inoculum was small the bacteria multiplied using the energy
derived from degrading the cellulose. The function to characterize and study the
bacterial response to Ca+2 concentrations on cellulose was also the logistic
model (Zwietering et al, 1990). Other models were discarded because the
substrate was only crystalline cellulose, whereby no rapid degradation or
disappearance occurs. Functions used normally to study nutrient degradation in
the rumen are first-order degradation kinetics Orskov and McDonald (1979) or
the model which includes a lag time, Orskov and McDonald (1979), modified by
Dhanoa (1988) (both cited by Nozziere and Michalet-Doreau, 2000 in their
review), but the data did not fit these models.
Thus, a non-linear regression analysis was used; the data were studied
using the model from Zwietering et al (1990) that describe a sigmoid curve. The
model is the same used for bacterial growth, the A, B and C parameters,
respectively, represent:
A: the maximum cellulose degradation, (mg cellulose)
B: the rate of cellulose degradation, (mg/h)
C: lag time (hours)
Also NLIN PROC from SAS was used for the analysis.
Due to the
different times when samples were taken because of the characteristic behaviors
of the different bacterial species, and the requirement from the analytical
program, it was not possible use NLIN MIXED for the analysis. All the species
and strains evaluated fitted the Zwietering model (P<0.05), with the exception of
R. albus 8.
Parameters A, B and C, and their respective standard errors, for each
tube were obtained, parameters were weighted by the reciprocal of the standard
error and analyzed by variance analysis (ANOVA) using PROC MIXED (SAS).
In the statistical model, Ca+2 concentrations, species, strain within species;
and the interaction: strain within species * Ca+2 concentrations were considered
as fixed effects; block and tube as random effects.
56
The extended model and the reduced model were compared. Since there
were no differences (P>0.05), the reduced model was use for ANOVA analysis.
LSM were compared by linear and quadratic contrasts (PROC MIXED, SAS).
Quadratic functions were estimated with PROC REG (SAS). Break point was
searched by maximum of the first derivate of quadratic function.
RA-8 did not fit the logistic model, in this case, linear regression was used,
at least to establish the rate of degradation (PROC REG, SAS); then the slopes
obtained were studied by ANOVA, but no significant differences were found
(P>0.05) among Ca+2 concentrations (GLM, SAS). The extent od degrdation
obtained at the last sampling was used to evaluated maximum degradation, the
data was analyzed by ANOVA and no differences were found (P>0.05).
The substrate affinity concept from Lineweaver-Burk method was also
applied to the rate of cellulose degradation (Price and Dwek, 1984), as was
made for growth experiments with Ca+2 and Mg (PROC REG, SAS).
Reduced model:
Y = [Ca+2] + species + strain (species)+ [Ca+2]*strain (species) + error
Y ijkl = Ci + Spj + St(sp)k:j+ C*Spij + C*St(sp)i k:j+ εijkl
Where: C ([Ca+2]),
Sp (Species),
i = 1, ….10
j = 1, 2, 3
St(sp) (Strain (species)), k = 1, 2
C*Sp, interaction Ca+2 with species
C*St(sp), interaction Ca+2 with Strain within Species
ε (Error),
i,j,k,l ( 0, σ2e )
57
Statistical analysis for cellulose degradation in NH3-free media with F.
succinogenes:
Similar to the other cellulose degradation experiments, the data were
studied using the model from Zwietering et al (1990), using NLIN PROC from
SAS. A, B and C parameters, respectively, represent:
A: the maximum cellulose degradation, (mg cellulose)
B: the rate of cellulose degradation, (mg/h)
C: lag time (hours)
Thus parameters A, B and C, and their respective standard errors, for
each tube were obtained, these were analyzed similarly to the other cellulose
experiment described above.
The statistical model included as fixed effects:
Ca+2 concentrations, strain; the interaction strain * Ca+2 concentrations and as
random effects: block and tube. The extended model and the reduced model
were compared; because no difference was found (P>0.05), the reduced model
was use for ANOVA analysis. Then LSM were analyzed by linear and quadratic
contrasts (PROC MIXED, SAS). Quadratic functions were estimated with PROC
REG (SAS). Break point was searched by maximum of the first derivate of
quadratic function.
In the case of FS-A3c NH3-free experiments the data from 0.02 mM Ca+2
did not fit the logistic model or the linear regression. Then LSM of cellulose
degraded per each sampling time were calculated and separated (DIFF option),
to establish the existence of differences in degradation among sampling times.
The substrate affinity concept from Lineweaver-Burk method was also
applied to the rate of cellulose degradation (Price and Dwek, 1984), as was
described earlier for growth experiments with Ca+2 and Mg and cellulose
degradation with different Ca+2 concentrations with normal N content.
58
Reduced statistical model:
Y = [Ca+2] + strain + [Ca+2]*strain + error
Y ijkl = Ci + Stj+ Bk + T(B)l:k + C*Stij + εijkl
Where: C ([Ca+2]),
St (Strain,
i = 1, ….7
k = 1, 2
C*St, interaction Ca+2 with Strain
ε (Error),
i,j,k,l,m ( 0, σ2e )
Statistical analysis of Ionized calcium and or magnesium requirements for
cellulose degradation, NH3-free cellulose media for R. flavefaciens
experiments.
In this experiment no degradation occurred with Ca+2-Mg free and Mg-free
and these treatmet were not included in the overall analysis; thus only the
treatments with Ca+2 + Mg (control) and Ca+2-free, were analyzed; then the
extent of degradation was estimated and the data for this parameter were
analyzed by ANOVA, obtaining LSM per strain*cation treatment interaction
(PROC GLM, SAS). No mean separation was made, because no differences
were found between treatments within strain (P>0.05).
The statistical model is:
Y = cation treatment + strain +strain*cation treatment+ error
Y ijk = Ti + Stj + St*Tij+ εijkl
Where: T (presence or absence of action),
St (Strain,
i = 1, 2
j = 1, 2
St*T (Strain*cation treatment interaction)
i,j,k ( 0, σ2e )
ε (Error),
59
CHAPTER 4
RESULTS AND DISCUSSION
4.1. Ionized calcium concentrations in cellobiose and cellulose media:
Ionized calcium concentrations in the media were regularly checked to
confirm that calculated and actual levels were similar (Table 4.1).
Later the
media were randomly analyzed, either to verify the concentrations or when was
necessary because experimental observations were markedly different than
expected (Figure 4.1).
Total Ca concentrations were higher than ionized calcium, as was
expected (Figure 4.2); Thus the regression Ca+2 = 0.5012*(total Ca) - 0.0237 (r2=
0.89) can be used to calculated Ca+2 concentrations from total Ca, and about
47% of total Ca is ionized Ca.
60
Measured
Calculated
Measured
Ca+2 (mM)
Ca+2 (mM)*
Calculated
total Ca
0.1M CaCl2
(ml added/
total Ca (mM)
(mM)**
liter medium)
0
0.0032-0.014
0.056
0
0
0.02
0.029
0.10
0.047
0.47
0.04
0.045
0.16
0.10
0.93
0.08
0.078
0.26
0.19
1.87
0.16
0.178
0.45
0.37
3.73
0.32
0.35
0.83
0.75
7.46
0.64
0.63
1.6
1.49
14.92
Normal
0.36
0.515
0.45
***
* Measured with an ion-electrode
** Measured by atomic absorption spectrophotometry
***0.33 g CaCl2/liter of mineral solution B (Appendix A)
Table 4.1. Ionized Ca concentration calculated and measured by ionelectrode, total Ca concentrations calculated and measured by atomic absorption
spectrophotometry.
61
Possible explainations for the differences between total Ca and Ca+2, even
though the source of calcium was CaCl2, which has 100% solubility in water
(Hodgman et al, 1959), are related to dilution and the presence of elements in
the anaerobic media that can precipitate or form complexes with Ca. These affect
the electrode response to the free calcium in solution, i.e., to ionized calcium
(Ca+2). Calcium can form soluble complexes with hydroxide, bicarbonate,
polyphosphates, citrate, tartrate and EDTA (ThermoOrion, 2001) and the extent
of this association of calcium is pH dependent: greater at higher calcium
concentrations and at higher pH values (pH:10-11) (ThermoOrion, 2001). NaOH
is used to adjust the medium pH to 6.7 to 6.9, and since the pH is neutral, the
tendency of hydroxide to form complexes with Ca+2 can be eliminated. Also Ca+2
can form relatively insoluble compound with oxalate, fluoride, phosphate and
sulfate and their solubility is increased at lower pH (ThermoOrion, 2001). Many
of these ions are present in the experimental anaerobic media (sulfate and
phosphate), and can potentially bind Ca+2. Due to the addition of a volatile fatty
acids (VFA) mix to the medium, its initial pH is aproximate 3.5, for this reason, to
avoid the possible complex formation at this pH, the CaCl2 was added to the
medium after pH was adjusted to 6.7 to 6.9 and after cysteineּHCl was added.
A background level of calcium in the medium was impossible to avoid,
since almost all reagent grade minerals contain trace amounts of calcium
(sodium carbonate, sodium chloride, magnesium sulfate).
Also the vitamin
Pantothenic acid, was only available as calcium pantothenate. In general the
quality of water was very good, with a few exceptions that were detected and
62
identified easily.
Desionized water was obtained from desionizer water
equipment (Nanopure DiamondTM), and quality was very similar to HPLC water
normally used as a reference.
Temperature is a factor affecting the solubility of different compounds
(Hodgman et al, 1959), and it is necessary to mention that the anaerobic media
was sterilized at 120º C, which could affect the solubilization of different
compounds and later the reconstitution of the different salts.
The pH can also affect the solubility of calcium salts in the media. As a
result of the fermentation that occurs in the culture tube, VFA are produced and
pH is decreased. For this reason, Ca+2 concentrations were verified under lower
pH attained by the addition of VFA (the same mix of VFA used to prepare the
media). Different amounts of VFA were added, pH was determined and Ca+2
measured, with no change in Ca+2 observed. The pH was decreased to 4.0,
lower than any culture condition, but still within the range of pH for normal
function of the ion electrode (as recommended by the manufacter, ThermoOrion,
2001).
63
Ca +2 measured
(mM)
1.2
y = 1.0032x + 0.0092
r2 = 0.99
1
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1
1.2
Ca+2 calculated (mM)
Figure 4. 1. Ca+2 concentrations in anaerobic medium measured by ion-electrode
vs. calculated (mM).
Ca +2 (mM)
0.8
y = 0.5012x - 0.0237
0.7
2
r = 0.89
0.6
0.5
0.4
0.3
0.2
0.1
0
0
Figure 4. 2.
0.5
1
1.5
total Ca (mM)
Ca+2 concentrations (measured with ion-electrode) vs. total Ca
concentrations (measured with AAS).
64
It is necessary to keep in mind the sensitivity and the limitations of the
determination of Ca+2 i.e., range of 10-5 to 10-1 M Ca+2
concentrations are
needed for a linear response. In addition, as mentioned earlier, the proper
conditions, such as temperature, pH, and lack of chemical interference from
other ions, can influence the proper analysis (ThermoOrion, 2001). Thus the
determinations of Ca+2 from Ca+2-free media, water, Ca+2-free mineral solution B
and mineral solution A can have more error than solutions or media with higher
Ca+2 concentrations.
4.2. Growth of predominant rumen cellulolytic bacteria on cellobiose media
with different Ca+2 concentrations.
In tables 4.2, 4.3 and 4.4 are presented the least squares means (LSM) of
the parameters used to characterize the growth curves of the bacteria grown with
different concentrations of Ca+2. These curves shown in figures 4.3, 4.4, 4.5,
and 4.7, are described by a logistic function with three parameters: maximum
growth defined as maximum optical density (max OD), rate of growth (OD
unit/hour) and lag time (hours). The statistical analysis by ANOVA identified as
significant sources of variation the main effects of species, strains within species,
Ca+2 concentrations, and the interactions of species*concentrations, and
concentrations*strain within species (all P<0.05) for max growth, rate of growth
and lag time. LSM of different Ca+2 concentrations were compared by linear and
quadratic contrasts.
65
Maximum growth for Fibrobacter succinogenes strain A3c (FS-A3c) did
not show neither linear or quadratic response to increased Ca+2 concentrations
(Table 4.2);
rate of growth increased in a linear (P=0.0027) and quadratic
(P=0.0075) manner to increased Ca+2 concentrations; and lag time had a linear
response (P=0.0084) to Ca+2 concentrations, decreasing the length of the lag
period when Ca+2 increased (Table 4.2, Figure 4.3).
Maximum growth of F.
succinogenes strain S85 (FS-S85) showed a response increasing linearly
(P=0.0012) and quadratically (P=0.0316) as Ca+2 concentrations increased; and
rate of growth increased linearly to increased Ca+2 concentrations (Table 4.2,
Figure 4.4). Neither a linear (P=0.1065) or quadratic (P=0.145) response to Ca+2
concentrations for lag time was observed for FS-S85.
For FS-A3c, Ca+2 requirements for growth cannot be estimated from the
actual results; the linear effect on rate of growth and lag time does not permit
reaching an end point, requiring the use of higher concentrations in future
experiments. Rate of growth had a quadratic response, and in this case, Ca+2
requirement was estimated as 0.43 mM, using the maximum of the first derivative
of the quadratic function (P=0.0001, r2=0.72).
FS-S85 responded linear and
quadratic to Ca+2 concentrations for max growth; then from the quadratic
response a requirement was estimated on 0.47 mM (using a quadratic function,
P=0.0205; r2=0.27).
From rate of growth, no requirement can be estimated
because the linear response observed.
Similar to FS-A3c higher Ca+2
concentrations will be needed in future research.
66
Max growth**
Rate of growth***
Lag time****
0*
0.02
0.08
0.16
0.32
0.36
0.64
1.43 ± 0.06
1.39 ± 0.04
1.37 ± 0.03
1.50 ± 0.04
1.40 ± 0.07
1.46 ± 0.03
1.51 ± 0.03
0.14 ± 0.018
0.15 ± 0.016
0.18 ± 0.016
0.18 ± 0.015
0.20 ± 0.025
0.20 ± 0.016
0.19 ± 0.016
17.2 ± 1.2
17.0 ± 1.1
15.0 ± 0.9
11.5 ± 0.9
17.4 ± 1.2
14.0 ± 0.9
11.1 ± 0.9
Linear1
Quadratic
0.1608
0.7988
0.0027
0.0075
0.0084
0.5403
0*
0.02
0.08
0.16
0.32
0.36
0.64
1.27 ± 0.06
1.27 ± 0.06
1.26 ± 0.04
1.37 ± 0.05
1.44 ± 0.08
1.34 ± 0.03
1.37 ± 0.04
0.11 ± 0.017
0.11 ± 0.016
0.13 ± 0.016
0.14 ± 0.015
0.13 ± 0.021
0.15 ± 0.016
0.14 ± 0.016
14.3 ± 1.1
15.2 ± 1.3
14.4 ± 0.9
13.9 ± 0.9
16.9 ± 1.2
13.3 ± 0.9
13.5 ± 0.9
Linear1
Quadratic
0.0012
0.0316
0.0039
0.2221
0.1065
0.1455
F. succinogenes A3c
F. succinogenes S85
* Ca+2, mM.
** OD units
*** OD units/hour
**** hours
1: for each strain, max growth, rate of growth and lag time were evaluated by
linear and quadratic contrasts, P values are presented.
Table 4.2. Effect of Ca+2 (mM) concentrations on growth of Fibrobacter
succinogenes strains A3c and S85 in liquid media with cellobiose as source of
carbohydrate.
67
OD
1.6
0
1.4
0.02
0.08
1.2
0.16
0.32
1
0.36
0.8
0.64
0.6
0.4
0.2
0
0
10
20
30
40
50 hours
Figure 4.3. Predicted growth curves of F. succinogenes A3c with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
1.6
1.4
0
1.2
0.02
1
0.08
0.16
0.8
0.32
0.6
0.36
0.64
0.4
0.2
0
0
10
20
30
40
50 hours
Figure 4.4. Predicted growth curves of F. succinogenes S85 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
68
Results on the effect of Ca+2 concentrations for Ruminococcus albus
strains 7 (RA-7) and 8 (RA-8) are shown in Table 4.3. For RA-7, maximum
growth did not show a linear or quadratic response to increased Ca+2
concentrations; however, there was a tendency for a linear response (P=0.0864).
Rate of growth of RA-7 had a tendency for a quadratic response (P=0.0565) to
increased Ca+2 concentrations, thus, RA-7 could reach the maximum rate of
growth at 0.36 mM Ca+2, according to the quadratic function estimation; but the
quadratic equation was not significant (P=0.2131, r2=0.12), then this estimation
lack precision. In table 4.3, the Ca+2 concentration of 0.36 mM shows the lower
rate of growth.
Lag times had a linear decrease with increasing Ca+2
concentrations (P=0.0065). The lack of response from RA-7 to increased Ca+2
concentrations is graphically shown in Figure 4.5, 4.8, and 4.9 where are shown
the predictive growth curves from logistic model, the curves of maximum growth
vs. Ca+2 concentrations, and the curves of rate of growth vs. Ca+2 concentrations,
respectively.
RA-8 grew very slowly and did not reach a max OD higher than 0.56, the
lowest among all the strains studied.
In addition, the growth curve showed
characteristics different from all other strains (Figure 4.6). This strain did not fit
the logistic function; therefore it was evaluated separately. Values for RA-8 were
adjusted to a linear function (P<0.05); the model and the slope were statistically
significant (P<0.05), but the intercept was not (P>0.05). When the rate of growth
represented by the slope of the linear functions, were compared among Ca+2
concentrations, RA-8 showed the highest rate of growth at 0.32 and 0.36 mM
69
Ca+2, with intermediary values for 0.08 and 0.64 mM Ca+2, while bacteria growing
with 0 and 0.02 mM Ca+2 showed the lowest rate of growth. Max OD reported in
table 4.3 corresponds to the observed values; lag time observed values had a
length of approximated to 20-22 hours, and time to reach the max OD ranged
118 to 211 hours.
Max growth**
Rate of growth***
Lag time****
R. albus 7
0*
0.02
0.08
0.16
0.32
0.36
0.64
1.21 ± 0.06
1.12 ± 0.04
1.11 ± 0.04
1.20 ± 0.06
1.24 ± 0.07
1.16 ± 0.03
1.23 ± 0.05
0.27 ± 0.021
0.25 ± 0.018
0.25 ± 0.016
0.26 ± 0.016
0.24 ± 0.028
0.23 ± 0.018
0.26 ± 0.017
7.5 ± 1.1
7.4 ± 1.2
7.1 ± 0.9
7.1 ± 0.9
7.1 ± 1.2
7.1 ± 0.9
6.7 ± 0.9
Linear1
Quadratic
0.0864
0.9869
0.1501
0.0565
0.0065
0.9654
R. albus 8
0*
0.02
0.08
0.16
0.32
0.36
0.64
0.46
0.48
0.51
0.50
0.55
0.55
0.53
(211)2
(199)
(159)
(206)
(118)
(127)
(173)
0.0020 ± 0.00018 a
0.0022 ± 0.00014 a
0.0031 ± 0.00023 b
0.0023 ± 0.00019 a
0.0044 ± 0.00029 c
0.0045 ± 0.00032 c
0.0029 ± 0.00024 b
20-223
* Ca+2, mM. ** OD units *** OD units/hour **** hours
1
: for each strain, max growth, rate of growth and lag time were evaluated by
linear and quadratic contrasts, P values are presented.
2
: max OD observed and time to attain it (hours)
3
: lag time observed (hours)
a,b,c: indicate differences (P<0.05) among Ca+2 concentrations for R. albus 8
Table 4.3. Effect of Ca+2 (mM) concentrations on growth of Ruminococcus albus
strains 7 and 8 in liquid media with cellobiose as source of carbohydrate.
70
OD
1.4
1.2
0
0.02
1
0.08
0.16
0.8
0.32
0.36
0.6
0.64
0.4
0.2
0
0
5
10
15
20
25
30
hours
Figure 4.5. Predicted growth curves of R. albus 7 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
0.6
0
0.02
0.5
0.04
0.4
0.64
0.3
0.2
0.1
0
0
50
100
150
200
Hours
Figure 4.6. Growth of R. albus 8 with different Ca+2 concentrations in cellobiose
media (max OD vs. time, raw data).
71
R. flavefaciens strains B34b (RF-B34b) and C94 (RF-C94) did not
respond to increased Ca+2 concentrations, in maximum growth, rate of growth or
lag time. Linear and quadratic contrast could not be calculated (P>0.05). The
results are presented in Table 4.4., and Figure 4.7 for RF-C94 due to the
particular high growth in Ca+2-free.
maximum growth with 0 mM
Consistently R. flavefaciens showed a
Ca+2, and this response was confirmed after
successive transfers from Ca+2-free media to Ca+2-free without affecting the
growth response (Appendix B).
72
Max growth**
Rate of growth***
Lag time****
R. flavefaciens B34b
0*
0.02
0.04
0.06
0.09
0.10
0.36
0.64
0.89 ± 0.06
0.88 ± 0.05
0.82 ± 0.19
0.86 ± 0.04
0.81 ± 0.04
0.88 ± 0.03
0.90 ± 0.03
0.85 ± 0.04
0.082 ± 0.017
0.077 ± 0.016
0.083 ± 0.045
0.080 ± 0.017
0.122 ± 0.016
0.096 ± 0.016
0.106 ± 0.017
0.093 ± 0.016
9.1 ± 1.3
7.3 ± 1.5
10.0 ± 1.2
8.0 ± 1.0
8.4 ± 1.0
7.4 ± 0.8
8.1 ± 1.0
6.8 ± 1.0
Linear1
Quadratic
NS
NS
NS
NS
NS
NS
R. flavefaciens C94
0*
0.02
0.04
0.06
0.09
0.10
0.36
0.64
0.86 ± 0.04
0.89 ± 0.05
0.63 ± 0.16
0.65 ± 0.04
0.69 ± 0.04
0.85 ± 0.04
0.82 ± 0.03
0.80 ± 0.05
0.162 ± 0.018
0.033 ± 0.016
0.024 ± 0.027
0.018 ± 0.016
0.022 ± 0.015
0.023 ± 0.015
0.054 ± 0.016
0.020 ± 0.015
18.9 ± 1.2
12.4 ± 1.7
12.2 ± 1.6
6.6 ± 1.2
5.3 ± 1.3
10.5 ± 0.9
10.7 ± 1.0
9.2 ± 1.1
Linear1
NS
NS
NS
Quadratic
NS
NS
NS
* Ca+2, mM.
** OD units *** OD units/hour **** hours
1: for each strain, max growth, rate of growth and lag time were evaluated by linear and
quadratic contrasts, P values are presented. NS= no solution.
Table 4.4. Effect of Ca+2 (mM) concentrations on growth of Ruminococcus
flavefaciens strains B34b and C94 in liquid media with cellobiose as source of
carbohydrate.
73
OD
1
0
0.9
0.02
0.8
0.04
0.7
0.06
0.6
0.09
0.1
0.5
0.36
0.4
0.64
0.3
0.2
0.1
0
0
10
20
30
40
50
hours
Figure 4.7. Predicted growth curves of R. flavefaciens C94 with different Ca+2
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
1.6
FS-A3c
1.4
FS-S85
1.2
RA-7
1
RA-8
RF-B34b
0.8
RF-C94
0.6
0.4
0.2
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Ca +2, mM
Figure 4.8. Curves of response to Ca+2 concentrations on maximum growth (OD)
by the different strains of rumen cellulolytic bacteria.
74
OD/hour
0.3
FS-A3c
FS-S85
0.25
RA-7
RA-8
RF-B34b
0.2
RF-C94
0.15
0.1
0.05
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Ca +2, mM
Figure 4.9. Curves of response to Ca+2 concentrations on rate of growth (OD
units/hour) by the different strains of rumen cellulolytic bacteria.
Considering the differences in response of the different species and
strains to increasing Ca+2 concentrations, it is important to evaluate the true
mineral requirements of these bacteria. Previous results on mineral requirement
reported by Bryant et al (1959) for FS-S85, have been considered to be universal
for all the rumen bacteria. They defined the requirements of FS-S85 for several
different minerals; among these, Ca and Mg, which were determined as a
function of max OD. They observed an increase in max OD with increasing
calcium concentrations.
Similarly, the present results indicate a quadratic
response for max OD by FS-S85. Bryan et al (1959) reported a requirement of
75
0.25 mM total Ca, which is lower than the requirement that was possible estimate
at the present study as 0.47 mM Ca+2, although this value may not very precise,
because the low r2 obtained (0.27). R. albus 7 showed a tendency for linear
response and for F. succinogenes A3c was not possible to establish a
requirement, due the lack of fit with different functions (linear, quadratic),
because the oscillatory response to different Ca+2 concentrations. Similarly for R.
flavefaciens, under the actual conditions no Ca+2 requirements for growth can be
determined.
When FS-A3c attained a max growth near to 1.4 OD in Ca+2-free media,
this result was unexpected, due to the observations from earlier experiments
where this bacterium did not growth in the Ca+2-free medium. The cultures were
examined microscopically and extremely pleomorphic cells were observed. On
this basis the Ca+2-free media used in this experiment was analyzed after the
experiment was completed, and found to contain 0.019 mM Ca+2. The other Ca+2
media also had higher concentrations than the estimated, probably because of
water contamination since the increase in concentration was similar.
In many experiments conducted during this study (not all reported), with
different objectives, (pre-experimental tests, response to different divalent
cations, consecutive bacterial transfers, etc), FS-A3c did not always grow in
Ca+2-free media (only 3 from a total 14 experiments).
In those experiments
where FS-A3c did not grow Ca+2 concentrations were determined as: 0.0138 mM
for Ca+2, and 0.053 mM for total Ca.
When a consecutive bacterial transfer
experiment was performed, transferring all the strains, from Ca+2-free to Ca+276
free media, when Ca+2 was 0.014 mM and total Ca was 0.046 mM, FS-A3c did
not grow (Appendix B).
However, when Ca+2 concentrations, from Ca+2-free
media, were 0.02 mM, FS-A3c grew normally until the 7th - 8th transfers, when
growth ceased. Obviously, this information confirms the small Ca+2 requirement
for FS-A3c. FS-S85, as well as all other organisms in this study always grew
under all the conditions described above. (Appendix B).
Using the Lineweaver-Burk method of plotting enzyme kinetic data (Price
and Dwek, 1984) to find the bacterial affinity for Ca+2 the reciprocal of the least
squares means of rate of growth of each treatment (Y) was plotted vs. the
reciprocal of Ca+2 concentrations (X), and a linear function was calculated. Only
data for FS-A3c and FS-S85 fit the linear functiona (Table 4.5). Then μmax
(maximum rate of growth) and Ks for each strain were calculated: 0.20 unit OD/h
and 0.0057 mM Ca+2, respectively, for FS-A3c, and 0.15 unit OD/h and 0.005
mM Ca+2, respectively, for FS-S85 (P<0.05).
For all the other strains, no
relationship between rate of growth with Ca+2 concentrations was found (P>0.05).
77
Linear function
Strain
P=
Intercept
Slope
r
2
μmax
Ks
(OD/h)
(mM)
FS-A3c
5.11±0.09 0.029±0.004 0.003 0.92 0.20
0.0057
FS-S85
7.04±0.22 0.036±0.01
0.005
Table 4.5.
0.03
0.75 0.15
Linear regression parameters from reciprocal of rate of growth vs.
reciprocal of Ca+2 concentrations, and μmax and Ks coefficients derivated from
the linear function.
Thus, it can be concluded that FS-S85 and FS-A3c, Gram (-) bacteria,
have a specific affinity with Ca+2, for growth process, and this is very low (0.005
and 0.0057 mM Ca+2). These concentrations may not be possible to attain under
our experimental conditions. R. albus strains 7 and 8 and R. flavefaciens strains
B34b and C94, (Gram (+)) did not show any substrate or Ca+2 affinity.
Bryant et al, (1959) demonstrated the essentiality of Ca for bacterial
growth in B. succinogenes S85, (later B. succinogenes was reclassified as F.
succinogenes) reporting that 0.05 mg /5 ml of media (0.25 mM Ca) was optimum
for growth. The authors did not confirm their Ca concentrations by mineral
analysis, assuming that the concentration was the amount added to the medium.
The present data suggest that the calcium requirements, for FS-S85 and FS-A3c
78
maybe higher than this concentration, and requirements for R. albus 7 may be
higher also; whereas no requirement could be demonstrated for R. flavefaciens.
Other studies that evaluated mineral requirements for bacteria from the
Bacteroides genus had shown different response to Ca concentrations. Thus,
Caldwell et al (1973) found no effect of Ca on growth of B. amylophilus (now
Ruminobacter amylophilus), and Caldwell and Arcand (1974) observed no effect
on B. ruminicola (now Prevotella ruminicola), but B. fragilis 2044 and B. oralis J1
were sensitive to the deletion of Ca from the growth media. The addition of Ca
(28 uM) to a cation-depleted media did not reestablish the normal growth of B.
ruminicola subsp. ruminicola B18, B. fragilis subsp. vulgatus 8483 and B.
succinogenes S85; but if Mg (10uM) was also added to the media, normal growth
was observed.
These results suggest differences among species in the
requirement for Ca. (Caldwell and Arcand, 1974) confirming the different
response from different strains to increased Ca+2 concentrations; although due to
the linear response it is not possible to determine the different requirements.
For a long time it was considered that Ca+2 did not play a metabolic role in
microbial metabolism (Silver, 1977) and that its function was relegated to the
outer (cell wall) environment of the bacteria. Durand and Kawashima (1980)
mention in their review that Ca is involved in the synthesis and stability of the cell
walls, and for this reason Gram (+) bacteria may have higher Ca and Mg
requirements for cell wall synthesis than Gram (-) bacteria. However, no clear
results were observed in the present study, where F. succinogenes (Gram (-))
and R. albus (Gram (+)) responded to Ca+2 concentrations,
79
whereas R.
flavefaciens (Gram (+)) did not, at least not in the same manner as the former
did. However, Beveridge et al (1982) showed that Gram (-) bacteria have lower
capacity to bind Ca+2 in the cell envelope, and thus they have a lower reservoir of
Ca+2 than Gram (+) bacteria. This could explain why Gram (-) bacteria are more
sensitive to Ca+2 concentrations in the media.
The different response from R.
albus and R. flavefaciens may be due to the cell wall composition of these
bacteria and their different capacity to bind different ions, such as calcium
(Beveridge, 1990, and Beveridge et al, 1985, 1986).
There is current evidence that Ca+2 acts as a signal under certain
environmental conditions that subsequently modify the internal concentration of
Ca+2 (Jones et al, 1999). Thus, Ca+2 can mediate the bacterial response as a
consequence of changes in osmotic or thermic conditions that drive different
processes such as sporulation, encapsulation, motility and reproduction (Ordal,
1977; Silver, 1977; Rampersaud et al, 1991; Tisa and Adler, 1995; Herbaud et al,
1998). It has been demonstrated that E. coli, a Gram (-) bacteria, requires Ca+2
to signal initiation of the reproduction process, activating the expression of a
specific protein, FtsZ, that signals the equator for cell division and allows division
to occur (Yu and Margolin, 1997; Onoda et al, 2000). For other bacteria, such as
Bacillus subtilis (Herbaud et al, 1998), the presence of Ca+2, affects both lag time
and activation of the reproduction process. When Ca+2 was present in the media,
the bacteria grew normally, whereas, in the absence of Ca+2, no growth was
observed.
80
E. coli and F. succinogenes share the characteristics of being Gram (-)
bacteria and perhaps the evidence available for E. coli is a key to explain the
observed Ca+2 response by F. succinogenes.
As was mentioned earlier, in
several experiments, FS-A3c grew in Ca+2-free medium. When the culture was
examined microscopically, (with the phase contrast microscope, immersion oil
objective) marked pleomorphism was observed.
This medium was later
analyzed for Ca+2 and found to contain 0.019 mM Ca+2. This observation was
found repeatedly when FS-A3c grew in low Ca+2 concentrations and is consistent
with concepts discussed above. Durand and Kawashima (1980) reported that Ca
deficiency in bacteria “can cause growth defects and the formation of
pleomorphic forms”, without any reference to the type, genus or species of
bacteria. When Ca+2 is deficient or absent, pleomorphism can be the result of the
abnormal function of the FtsZ protein that is responsible for the polymerization of
the enzoskeleton, which in turn signals the equator to unsergo cell division. In the
absence of Ca+2, E. coli can grow but does not divide and forms long
multinucleated filaments (Yu and Margolin, 1997; Onoda et al, 2000). Onoda et al
(2000) reported that 1.2 mM of Ca+2 was necessary to reestablish normal growth.
4.3. Cations required for bacterial growth with specific attention to
magnesium requirements.
Since R. flavefaciens did not require Ca+2 at a concentration above the
background in the basal medium and sequential transferring from Ca+2-free
media to Ca+2-free media did not decrease max OD, the absolute requirements
81
for other specific divalent cations (Fe, Mg, Mn and Zn) were investigated. Both
strains of R. flavefaciens had an absolute requirement for Mg (Table 4.6.), since
any combinations of divalent cations without Mg, did not support growth.
R.
albus 7 had a decrease in max growth when Mg was not present in the media,
alltough it was able to growth in divalent cation free media (morphological
changes were observed under the microscope i.e., agglutination of cells and the
loss of coccus form). F. succinogenes A3c did not grow when Ca+2 was omitted
from the different combinations of divalent cation;
however, the absence of
either Mg, Zn, Fe, or Mn permitted a normal growth. In contrast, FS-S85 was not
affected by the individual absence of Ca+2 or any divalent cation tested, but did
not grow or only grew very slowly when Ca+2 and Mg were both omitted from the
medium.
82
Organism
Medium
Normal
Divalent cation free
Ca+2-Mg free
Ca+2-free
Mg-free
Zn-free
Mn-free
Fe-free
+Ca+2
+Mg
+Zn
+Mn
+Fe
+Ca+2+Mg
+Ca+2+Zn
+Ca+2+Mn
+Ca+2+Fe
+Mg+Mn
+Mg+Zn
+Mg+Fe
+Zn+Mn
L+Zn+Fe
+Ca+2+Mg+Zn
FS-S85
FS-A3c
N
-/L
-/L
N
-/N
N
L/N
N
L
N
N
N
L
-/N
N
N
-/L
N
L
N
N
N
N
-/L
N
N
-/L
N
L
N
N
N
N
L
N
RF-B34b
N
N
N
N
N
N
L
L
-
L
L
-
L
L
L
L
L
L
L
L
L
L
RA-7
N
L
L
N
L
N
N
L
L
N
L
N
L
N
L
L
L
N
L
L
L
N
N
L
L
L
N
L
N
L
L
L
N
L
L
L
N
N
L
L
L
N
(-) : negative, no growth;
N: normal growth and L: lower growth than N (75% or lower than max OD of N).
Each column represents an experiment in duplicate, when a different response
was observed between duplicates the response for each one is indicated (i.e., /N or -/L).
Table 4.6. Growth response to divalent cations in cellobiose liquid medium by
predominant rumen cellulolytic bacteria.
83
Since all the bacterial species used in the Ca+2 experiments were affected
by omission of Mg from the medium their requirement for Mg was investigated.
Treatments evaluated only total concentration of Mg, because no methods were
available to measure ionized Mg concentrations.
Growth, measured as OD, was characterized by a logistic function, which
is similar to that for the results from Ca+2 treatments. The statistical analysis of
the parameters of the growth curves showed significant main effects for species,
strain
within
species,
Mg
concentrations
and
for
the
interactions:
species*concentration and strain within species*concentrations (P< 0.05). Since
RF-B34b and RF-C94 did not grow when they were cultured in Mg-free media,
the data from 0 Mg concentration was not included in the statistical analysis.
As can be seen in Table 4.7 and Figure 4.10 and 4.11, neither RF-B34b or
RF-C94 grew in Mg-free medium. Maximum growth for RF-B34b responded to
both linear and quadratic contrasts (P= 0.0009 and 0.0138, respectively), and
RF-C94 responded linearly increasing max OD (P=0.0010) to increased Mg
concentrations.
Rate of growth differed with Mg concentrations; RF-B34b
responded to quadratic contrast (0.0020) whereas RF-C94 had a linear response
(0.0532). Lag time of R. flavefaciens was sensitive to Mg concentrations, thus
longer lag times were observed for both strains at lower Mg concentrations,
showing a linear and quadratic response (P<0.05). Thus, using the maximum of
the quadratic function (Appendix C), Mg requirements can be estimated for RFB34b as ~0.5 mM of Mg, considering max growth/max OD or rate of growth as
84
parameters.
RF-C94 showed a increased linear response to increased Mg
concentrations, therefore requirement could not be determined because an end
point or plateau was not attained in the present study.
Max growth**
Rate of growth***
Lag time****
R. flavefaciens B34b
0*
0.05
0.1
0.2
0.41
0.82
No growth
0.50 ± 0.07
0.62 ± 0.03
0.76 ± 0.04
0.88 ± 0.05
0.80 ± 0.04
0.045 ± 0.002
0.073 ± 0.005
0.099 ± 0.006
0.085 ± 0.006
0.089 ± 0.009
16.5 ± 1.00
12.4 ± 0.75
10.4 ± 0.68
11.4 ± 0.67
9.3 ± 0.71
Linear1
Quadratic
0.0009
0.0138
0.4341
0.0020
0.0004
0.0030
R. flavefaciens C94
0*
0.05
0.1
0.2
0.41
0.82
No growth
0.25 ± 0.03
0.51 ± 0.04
0.52 ± 0.03
0.53 ± 0.03
0.62 ± 0.03
0.011 ± 0.001
0.025 ± 0.003
0.024 ± 0.003
0.028 ± 0.004
0.029 ± 0.004
25.0 ± 1.45
15.4 ± 0.93
12.8 ± 0.98
10.2 ± 1.02
11.6 ± 0.92
Linear1
Quadratic
0.0010
0.1631
0.0532
0.1403
0.0014
0.0042
*Mg, mM.
** OD units *** OD units/hour **** hours
1
: for each strain, max growth, rate of growth and lag time were evaluated by
linear and quadratic contrasts, P values are presented.
Table 4.7. Growth of Ruminococcus flavefaciens strains B34b and C94 with
different concentrations of Mg in liquid media.
85
OD
1
0.9
0
0.8
0.05
0.7
0.1
0.6
0.2
0.5
0.41
0.4
0.82
0.3
0.2
0.1
0
0
10
20
30
40
50
60
hours
Figure 4.10. Predicted growth curves of R. flavefaciens B34b with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
0.7
0.6
0
0.5
0.05
0.1
0.4
0.2
0.41
0.3
0.82
0.2
0.1
0
0
8
16
24
32
40
48
56
64
hours
Figure 4.11. Predicted growth curves of R. flavefaciens C94 with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
86
Maximum growth of R. albus 7 was reduced when Mg was not present in
the media (Table 4.8. Figure 4.12). This strain was also able to growth in divalent
cation free media (although morphological changes were observed). Max growth
for R. albus 7 responded linearly (P=0.0568) to increasing Mg concentrations,
with a tendency for a quadratic response (P=0.0744). No effect on rate of growth
or lag time was observed for RA-7 with increasing Mg concentrations (linear and
quadratic contrasts were not solved).
R. albus 8 (Table 4.8; Figure 4.13), showed a quadratic response
(P=0.0154) for max growth (max OD) when Mg concentrations increased,
estimating the Mg requirement for maximum growth as 0.5 mM Mg (using the
maximum of the quadratic function, appendix C).
Data for RA-8 for rate of
growth and lag time did not show structure for either a linear or quadratic
contrasts.
Then, RA-7 showed, apparently, to have higher Mg requirements for
maximum growth than RA-8 (estimated as 0.5 mM Mg).
Thus, differences
between the two R. albus strains for Mg requirements are evident from the
present results.
87
Max growth**
Rate of growth***
Lag time****
R. albus 7
0*
0.05
0.1
0.2
0.41
0.82
0.80 ± 0.06
1.05 ± 0.06
1.19 ± 0.06
1.21 ± 0.06
1.19 ± 0.06
1.19 ± 0.04
0.16 ± 0.027
0.22 ± 0.005
0.24 ± 0.010
0.25 ± 0.012
0.24 ± 0.014
0.28 ± 0.019
6.5 ± 0.83
6.9 ± 0.72
7.1 ± 0.68
6.8 ± 0.68
7.0 ± 0.68
6.8 ± 0.77
Linear1
Quadratic
0.0568
0.0744
NS
NS
NS
NS
R. albus 8
0*
0.05
0.1
0.2
0.41
0.82
0.53 ± 0.04
0.47 ± 0.04
0.49 ± 0.04
0.58 ± 0.04
0.59 ± 0.03
0.55 ± 0.04
0.003 ± 0.002
0.004 ± 0.001
0.004 ± 0.001
0.005 ± 0.001
0.006 ± 0.002
0.005 ± 0.002
12.5 ± 2.44
15.5 ± 2.41
13.8 ± 2.27
23.0 ± 2.03
15.5 ± 1.95
36.1 ± 2.03
Linear1
Quadratic
0.1588
0.0154
NS
NS
NS
NS
*Mg, mM.
** OD units *** OD units/hour **** hours
1
: for each strain, max growth, rate of growth and lag time were evaluated by
linear and quadratic contrasts, P values are presented.
Table 4.8. Growth of Ruminococcus albus
concentrations of Mg in liquid media
88
strains 7 and 8 with different
OD
1.4
1.2
0
1
0.05
0.1
0.8
0.2
0.41
0.6
0.82
0.4
0.2
0
0
4
8
12
20 hours
16
Figure 4.12. Predicted growth curves of R. albus 7 with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
0.7
0
0.6
0.05
0.5
0.1
0.2
0.4
0.41
0.3
0.82
0.2
0.1
0
0
50
100
150
200
250
300
hours
Figure 4.13. Predicted growth curves of R. albus 8 with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
89
Both strains of F. succinogenes responded in max growth to increased
Mg concentrations (Table 4.9., Figures 4.14 and 4.15). FS-A3c had a positive
linear response (P=0.0366) to increased Mg concentrations, whereas FS-S85
showed, both, increasing max growth in linear and quadratic way (P<0.0001,
respectively) to increased Mg concentrations. Thus, for both strains the lower
maximum growth was obtained with the lower Mg concentration. FS-A3c also
had a positive linear response to Mg concentrations for rate of growth (P=0.0027)
and only a tendency to decrease the lag time linearly (P=0.0725) as Mg
concentration increased. The lower rate of growth and the longer lag time was
observed with 0 Mg for FS-A3c.
FS-S85 did not response to increased Mg
concentrations, neither linearly or quadratic, for rate of growth and lag time.
Thus, Mg requirement for FS-A3c cannot be defined from the data available and
the method used for it estimation (Figure 4.16 and 4.17).
For FS-S85, Mg
requirements for maximum growth can be estimated as 0.56 mM of Mg from the
quadratic function (Appendix C). Apparently, FS-A3c has a higher requirement
for Mg than FS-S85, considering that
FS-A3c had a linear response to Mg
concentrations. The requirement for FS-S85 attained in this study is higher than
that reported by Bryant et al (1959) for the same strain (0.1 mM Mg).
90
Max growth**
Rate of growth***
Lag time****
F. succinogenes A3c
0*
0.05
0.1
0.2
0.41
0.82
0.99 ± 0.14
1.43 ± 0.11
1.54 ± 0.06
1.54 ± 0.07
1.53 ± 0.05
1.63 ± 0.07
0.12 ± 0.026
0.17 ± 0.007
0.18 ± 0.006
0.18 ± 0.009
0.20 ± 0.010
0.17 ± 0.010
15.3 ± 1.11
11.5 ± 1.29
12.4 ± 0.64
12.0 ± 0.73
11.9 ± 0.62
12.7 ± 0.63
Linear1
Quadratic
0.0366
NS
0.0027
NS
0.0725
NS
F. succinogenes S85
0*
0.05
0.1
0.2
0.41
0.82
0.86 ± 0.06
1.06 ± 0.05
1.24 ± 0.05
1.34 ± 0.05
1.29 ± 0.05
1.35 ± 0.05
0.05 ± 0.017
0.14 ± 0.004
0.14 ± 0.007
0.15 ± 0.008
0.15 ± 0.009
0.15 ± 0.011
17.1 ± 0.94
15.3 ± 0.77
14.8 ± 0.75
14.7 ± 0.72
15.0 ± 0.72
14.3 ± 0.72
Linear1
Quadratic
<0.0001
<0.0001
NS
NS
NS
NS
* Mg, mM. ** OD units *** OD units/hour **** hours
1
: for each strain, max growth, rate of growth and lag time were evaluated by
linear and quadratic contrasts, P values are presented.
Table 4.9. Growth of Fibrobacter succinogenes strains A3c and S85 with different
concentrations of Mg in liquid media.
91
OD
1.8
1.6
0
1.4
0.05
1.2
0.1
0.2
1
0.41
0.8
0.82
0.6
0.4
0.2
0
0
8
16
24
32
40
hours
Figure 4.14. Predicted growth curves of F. succinogenes A3c with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
OD
1.6
1.4
0
1.2
0.05
0.1
1
0.2
0.8
0.41
0.6
0.82
0.4
0.2
0
0
10
20
30
40
50
60
hours
Figure 4.15. Predicted growth curves of F. succinogenes S85 with different Mg
concentrations in cellobiose media, using the parameters obtained from the
logistic function.
92
OD
1.8
FS-A3c
1.6
FS-S85
1.4
RA-7
1.2
RA-8
1
RF-B34b
0.8
RF-C94
0.6
0.4
0.2
0
Mg, mM
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Figure 4.16. Curves of response to Mg concentrations on maximum growth (OD)
by the different strains of rumen cellulolytic bacteria.
OD/hour
0.3
FS-A3c
FS-S85
0.25
RA-7
RA-8
RF-B34b
0.2
RF-C94
0.15
0.1
0.05
0
Mg, mM
0
0.2
0.4
0.6
0.8
1
Figure 4.17. Curves of response to Mg concentrations on rate of (OD units/hour)
by the different strains of rumen cellulolytic bacteria.
93
The same approach used with Ca+2 and substrate affinity was applied to
evaluate Mg affinity by the bacteria used in the present study. Linear functions
were obtained for all the strains (Table 4.10) with the exception of FS-A3c, which
did not fit the linear function (P>0.05). Values calculated, μmax (units OD/h) and
Ks (mM Mg), from these functions were respectively, for RF-B34b: 0.11 and
0.067; RF-C94: 0.038 and 0.104; RA-7: 0.27 and 0.01; RA-8: 0.0056 and 0.031;
FS-S85: 0.15 and 0.from highest to lowest: FS-S85 > RA-7 > RA-8 > RF-B34b >
RF-C94. This is consistent with the higher Mg requirements observed for R.
flavefaciens.
Linear function
Strain
P=
r2
μmax
Ks
(OD/h)
(mM)
Intercept
Slope
RF-B34b
9.12± 1.28*
0.61± 0.12*
0.016 0.89
0.11
0.067
RF-C94
26.69± 6.61*
2.77± 0.64*
0.023 0.86
0.038
0.104
RA-7
3.76 ± 0.15*
0.038± 0.014**
0.079 0.70
0.27
0.01
RA-8
177.3± 14.3*
5.44 ± 1.39*
0.029 0.84
0.0056
0.031
FS-S85
6.58± 0.1*
0.039± 0.01*
0.027 0.85
0.15
0.006
* P<0.05
Table 4.10.
** 0.05<P>0.10
Linear regression parameters from reciprocal of rate of growth vs.
reciprocal of Mg concentrations, and μmax and Ks coefficients derivated from the
linear function.
94
Thus, only R. flavefaciens has an absolute requirement for Mg, while the
remaining strains have different Mg requirements for optimum growth.
Several different authors (Bryant et al, 1959; Scott and Dehority, 1965,
Caldwell et al, 1973; Caldwell and Arcand, 1974) have used maximum growth
(max OD) as the indicator for determining nutrient requirements. Max OD was
very useful in determining Mg requirement in the present study, whereas this was
not the case with Ca+2. Possibly, a consideration of the different roles of these
two cations in bacterial metabolism may provide clues as how to requirements
are influenced and how they can be measured. Because Mg functions in so
many different processes in the cell, i.e., associated with ribosomes, as
coenzyme in CHO metabolism, synthesis and integrity of cell walls (Jasper and
Silver, 1977; Durand and Kawashima, 1980; Smith and Maguire, 1998). On the
other hand, Ca+2 appears more like a signaling molecule (Herbaud et al, 1998).
Thus max growth may be a better reflection of Mg requirements, because the
limiting nutrient is affecting both rate and extent of bacterial growth.
Considering the role of Mg in cell wall synthesis, it could be logical to
expect a high Mg requirement for Gram (+) bacteria, as was reported by Durand
and Kawashima (1980); and observed for R. flavefaciens, which has an absolute
requirement for Mg.
Apparently R. albus has lower requirements than R.
flavefaciens; at least R. albus can growth without Mg in the medium. This may
be due to differences in metabolic demands, different capacity to control
intracellular concentrations or Mg-channels may be different between the two
species. Unfortunately, all of these explanations are only speculations, due to the
95
lack of information related to this topic in rumen bacteria. Information from
Salmonella, E. coli and other pathogens shows differences in resistance to Mg
starvation, which may be due to different channel systems and variations in how
different species manage their Mg status (Jasper and Silver, 1977; Smith and
Maguire, 1998).
Beveridge (1990) and Beveridge et al (1985, 1986) reported differences in
cell wall composition between two species of Bacillus, B. subtilis, and B.
licheniformis. These differences consisted of changes in the amount and
structural disposition of teichoic, lipoteichoic and teicuronic acid in the cell wall
matrix, which altered the capacity of these bacteria to bind different ions and in
different proportions. The differences in Mg affinity observed for the predominant
rumen cellulolytic bacteria may be related with differences in their cell wall
composition, but the information available about cell wall composition for these
bacteria is not enough to establish this different capacity to bind Mg (Dehority,
1977; Vinogradov et al, 2001).
4.4. Cellulose degradation and Ca+2 concentrations:
The results for cellulose degradation with varying Ca+2 concentrations are
presented in Tables 4.11, 4.12 and 4.13 as LSM of the parameters: maximum
extent of degradation, rate of degradation and lag time. The statistical analysis
showed significant (P<0.05) main effects for species, strain within species, Ca+2
concentration and for interactions of species*Ca+2 concentration and strain within
96
species*Ca+2 concentration, for maximum degradation and lag time. For rate of
degradation all main effects (P<0.05), with the exception of Ca+2 concentrations
(P>0.05), and all measured interactions were significant (P<0.05).
R. albus 8 did not fit the logistic function. Maximum degradation was
estimated from the data of degradation at last sampling time and arithmetic
average is reported for each Ca+2 concentration, and analyzed by ANOVA
(PROC GLM, SAS). To estimate rate of degradation the data were analyzed by
linear regression (P<0.05); the slopes obtained later were analyzed by ANOVA
(PROC GLM, SAS) and LSM per Ca+2 concentration were obtained. However no
effect of Ca+2 concentrations on rate of cellulose degradation by RA-8 was found
(P>0.05).
The maximum cellulose degradation and lag time were affected by Ca+2
concentrations in all the strains, except for RA-8. Rate of degradation was not
affected by Ca+2 concentrations (P>0.05). Interactions of species*treatment and
strain within species*treatment were detected significant (P<0.05); thus, only F.
succinogenes A3c and S85, showed that rate of cellulose degradation was
affected by the Ca+2 concentrations (P<0.05).
Neither strain of F. succinogenes degraded cellulose in the absence of
supplemental Ca+2; however, they both responded to increased Ca+2
concentrations with an irregular pattern of maximum cellulose degradation (table
4.11, figure 4.18, 4.19, and 4.24). FS-A3c had a linear response for maximum
degradation and rate of degradation (P<0.0001), meaning that higher Ca+2
concentrations need be tested to establish their requirement. Although there is
97
an effect of Ca+2 concentrations on maximum degradation for FS-S85, it cannot
be represented by a linear or quadratic adjustment, the contrasts analysis had no
solution (P>0.05). In figure 4.24, for both strains, maximum cellulose degradation
plotted vs. Ca+2 concentrations shows the incapacity to reach a break point and
then define a requirement. It would be necessary to test higher concentrations
than those used in this experiment. Although, lag time was significant affected
by Ca+2 concentrations, it did not fit the linear and quadratic contrasts for FS-A3c
(P=0.6165, NS) and FS-S85 (no solution for contrasts).
98
Max
degradation**
Rate of
degradation***
Lag time****
F. succinogenes A3c
0*
0.04
0.06
0.08
0.12
0.36
No degradation
21.0 ± 0.26
14.0 ± 0.07
21.5 ± 0.09
20.6 ± 0.11
32.0 ± 0.07
0.33 ± 0.18
0.99 ± 0.11
0.69 ± 0.29
0.51 ± 0.20
2.00 ± 0.08
49.0 ± 6.24
37.5 ± 2.78
35.5 ± 2.87
16.4 ± 4.03
28.8 ± 2.72
Linear1
Quadratic
<0.0001
NS
<0.0001
NS
0.6165
NS
F. succinogenes S85
0*
0.04
0.06
0.08
0.12
0.36
No degradation
25.0 ± 0.09
22.2 ± 0.09
28.0 ± 0.08
25.8 ± 0.09
31.8 ± 0.11
1.11 ± 0.18
1.24 ± 0.37
2.48 ± 0.28
1.25 ± 0.46
1.45 ± 0.25
20.4 ± 2.92
21.0 ± 2.82
22.2 ± 2.75
23.4 ± 2.81
21.2 ± 3.05
Linear1
Quadratic
NS
NS
NS
NS
NS
NS
*Ca+2 mM
**mg cellulose ***mg cellulose/hour
****hours
1
: for each strain, maximum degradation, rate of degradation and lag time were
evaluated by linear and quadratic contrasts, P values are presented.
Table 4.11. Cellulose degradation by Fibrobacter succinogenes strains A3c and
S85 incubated with different concentrations of ionized Ca+2.
99
mg
35
30
0
0.04
25
0.06
0.08
20
0.12
0.36
15
10
5
0
0
10
20
30
40
50
60
70
80
hours
Figure 4. 18. Predicted cellulose degradation curves of F. succinogenes A3c with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function.
mg
35
30
0
0.04
25
0.06
20
0.08
0.12
15
0.36
10
5
0
0
10
20
30
40
50
60
70
80 hours
Figure 4. 19. Predicted cellulose degradation curves of F. succinogenes S85 with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function.
100
The highest maximum degradation among all the strains studied was
obtained with RA-7, which appears to be the most aggressive cellulolytic strain
(Table 4.12 and Figure 4.20). Ca+2 concentrations had a significant effect on
cellulose degradation; RA-7 responded with increased maximum degradation
either
linearly
(P=0.0027)
or
quadratic
(P=0.0372)
to
increased
Ca+2
concentrations. The lowest maximum degradation was obtained with 0 Ca+2,
increasing when Ca+2 increased, showing a tendency to reach a plateau with
higher concentrations (Figure 4.24);
RA-7 responded decreasing lag time,
linearly or quadratic, to increased Ca+2 concentrations. The requirement for
maximum degradation and lag time were estimated as 0.28 and 0.34 mM Ca+2,
respectively.
The quadratic functions used to estimate the requirement for
maximum degradation, lack of significance (P=0.1440; r2=0.12); only the function
for lag time was significant (P<0.0001; r2=0.54), then be the one useful for
estimation; although lag time is not a parameter commonly used to evaluate
requirements.
RA-8 was analyzed independently, since its values did not fit the logistic
function (Figure 4.21). The maximum degradation, averages of the extent of
degradation obtained at the last sampling, were analyzed by ANOVA, but no
effect of Ca+2 concentrations was detected (P>0.05). The slope of the linear
function for each Ca+2 concentration, representing the rate of cellulose
degradation, was analyzed by ANOVA and did not show differences due the
effect of Ca+2 concentration, thus no further analysis was made (P>0.05) (Table
4.12.). Because of the lack of fit with the logistic function, the parameter “C”,
101
representing lag time, was not obtained.
In table 4.12 the time of the first
sampling was reported as lag time, only as a reference for the time delay until
evident cellulose degradation was observed and samples were taken.
Max Cellulose
degradation**
Rate of
degradation***
Lag time****
R. albus 7
0*
0.02
0.04
0.08
0.16
0.32
0.36
0.64
31.34 ± 1.37
38.32 ± 1.45
36.59 ± 1.46
34.30 ± 1.59
37.88 ± 1.53
36.94 ± 1.46
33.46 ± 1.74
33.96 ± 1.76
1.08 ± 0.19
1.41 ± 0.26
1.48 ± 0.28
1.49 ± 0.35
1.50 ± 0.31
1.47 ± 0.29
1.23 ± 0.36
1.55 ± 0.41
15.62 ± 4.02
10.90 ± 4.11
12.11 ± 4.13
10.61 ± 4.29
9.36 ± 4.22
9.84 ± 4.16
7.09 ± 4.55
12.57 ± 4.42
Linear1
Quadratic
0.0027
0.0372
N.C.
<0.0001
0.0009
R. albus 8
9.802
0*
0.02
15.10
0.04
13.55
0.08
19.68
0.16
18.83
0.32
26.00
0.36
21.96
0.64
21.10
+2
* Ca mM **mg of cellulose
0.0285 ± 0.0163
0.0418 ± 0.017
0.0589 ± 0.043
0.0395 ± 0.028
0.0574 ± 0.021
0.0854 ± 0.026
0.0659 ± 0.021
0.0649 ± 0.022
***mg cellulose/hour ****hours
604
60
60
60
60
60
60
60
1
: for R. albus 7, max degradation and lag time were evaluated by linear and quadratic
contrasts, P values are presented. N.C. : No contratsts analysis, no effect of Ca+2
concentrations (P>0.05).
For R. albus 8, for max cellulose degradation (2) extent of degradation obtained at the
last sampling time (275 hours), for rate of degradation (3) the slope of the linear
regression (P=0.064), for lag time (4) the time for first sampling determined visually.
Table 4.12. Cellulose degradation by Ruminococcus albus strains 7 and 8
incubated with different concentrations of Ca+2.
102
mg
45
40
0
35
0.02
30
0.04
0.08
25
0.16
20
0.32
15
0.36
0.64
10
5
0
0
10
20
30
40
50
60
70
80
hours
Figure 4. 20. Predicted cellulose degradation curves of R. albus 7 with different
Ca+2 concentrations in cellulose media, using the parameters obtained from the
logistic function.
g degraded
0.045
Ca-free
0.04
0.02
0.035
0.04
0.08
0.03
0.16
0.32
0.025
0.36
0.02
0.64
0.015
0.01
0.005
0
hours
0
50
100
150
200
250
300
Figure 4. 21. Cellulose degradation curves of R. albus 8 with different Ca+2
concentrations in cellulose media (raw data data from replicate 1).
103
Extent of cellulose degradation and lag time for R. flavefaciens B34b and
C94 were both affected by Ca+2 concentrations (P<0.05) (Table 4.13 and Figure
4.22
and 4.23). Rate of degradation was not affected by Ca+2 concentrations
(P>0.05). For RF-B34b and RF-C94, maximum degradation increased linearly
(P<0.0001) as Ca+2 concentrations increased. However, this response should be
taken with care, since it may be an artifact of the adjustment
Figure 4.24
graphically shows the tendency to increase degradation, but with variations that
do not permit identifying a break point. The linear function fitted the adjustment,
although it does not reflect the biology of the response. For both strains, lag time
was affected by Ca+2 concentrations, but only RF-C94 showed a quadratic
response (P=0.0007) to the contrasts evaluation.
The estimated quadratic
function lack of significance (P=0.2659; r2=0.06), then the calculated Ca+2
requirement 0.1 mM is not correct.
However, lag time is not a common
parameter used to measure requirements.
104
Max Cellulose
degradation**
Rate of
degradation***
Lag time****
R. flavefaciens B34b
0*
0.02
0.04
0.08
0.16
0.32
0.36
0.64
18.24 ± 1.10
16.87 ± 1.22
17.25 ± 1.19
19.24 ± 0.94
19.27 ± 0.95
20.83 ± 1.01
18.90 ± 1.23
21.44 ± 1.17
0.885 ± 0.30
0.479 ± 0.20
0.542 ± 0.18
0.729 ± 0.16
0.776 ± 0.18
0.743 ± 0.17
0.734 ± 0.23
0.708 ± 0.21
15.94 ± 3.78
9.14 ± 4.05
11.02 ± 3.75
12.61 ± 3.40
12.87 ± 3.42
11.90 ± 3.45
15.89 ± 3.67
11.42 ± 3.69
Linear1
Quadratic
<0.0001
0.3507
N.C.
0.9585
0.9850
R. flavefaciens C94
0*
0.02
0.04
0.08
0.16
0.32
0.36
0.64
26.43 ± 1.58
28.58 ± 1.13
28.78 ± 1.60
30.23 ± 1.47
28.51 ± 1.49
35.64 ± 1.53
26.83 ± 1.74
27.74 ± 1.64
0.994 ± 0.23
0.742 ± 0.19
0.462 ± 0.25
0.742 ± 0.29
0.579 ± 0.25
1.006 ± 0.35
0.534 ± 0.26
0.464 ± 0.23
15.51 ± 4.57
23.62 ± 3.59
8.83 ± 4.86
8.46 ± 4.67
8.83 ± 4.79
14.36 ± 4.37
24.33 ± 4.60
6.74 ± 4.95
Linear1
Quadratic
<0.0001
0.1222
N.C.
0.1955
0.0007
* Ca+2 mM
**mg of cellulose
***mg cellulose/hour
****hours
N.C. : No contratsts analysis, no effect of Ca+2 concentrations (P>0.05).
Table 4.13. Cellulose degradation by Ruminococcus flavefaciens strains B34b
and C94 incubated with different concentrations of Ca+2.
105
mg
25
0
0.02
20
0.04
0.08
0.16
15
0.32
0.36
10
0.64
5
0
0
10
20
30
40
50
60
70
80
hours
Figure 4. 22. Predicted cellulose degradation curves of R. flavefaciens B34b with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function.
mg
40
0
0.02
35
0.04
30
0.08
0.16
25
0.32
20
0.36
15
0.64
10
5
0
0
10
20
30
40
50
60
70
80
hours
Figure 4. 23. Predicted cellulose degradation curves of R. flavefaciens C94 with
different Ca+2 concentrations in cellulose media, using the parameters obtained
from the logistic function.
106
mg degraded
45
FS-A3c
FS-S85
40
RA-7
35
RA-8
RF-B34b
30
RF-C94
25
20
15
10
5
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Ca +2, mM
Figure 4.24. Curves of response to Ca+2 concentrations on maximum cellulose
degradation (mg degraded) by the different strains of rumen cellulolytic bacteria.
mg/hour
3
FS-A3c
FS-S85
RA-7
2.5
RF-B34b
RF-C94
2
1.5
1
0.5
+2
0
Ca , mM
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Figure 4.25. Curves of response to Ca+2 concentrations on rate of degradation
(mg degraded/hour) by the different strains of rumen cellulolytic bacteria.
107
Similar to earlier experiments, the kinetic approach was used to study the
response from FS-A3c and FS-S85 to Ca+2 treatments on cellulose degradation;
Only F. succinogenes strains were included since only they presented different
rates of degradation due to different Ca+2 concentrations (identified from the
significant interaction (P<0.05), although, they are not protective by the effect of
Ca+2 concentrations). Linear functions were obtained, but the statistical
significance showed only a tendency (Table 4.14).
Linear function
Strains
P=
r
2
μmax
Ks
(mg/h)
(mM)
intercept
slope
FS-A3c
0.50± 0.47
0.102±0.03
0.088
0.83
2.01
0.204
FS-S85
0.69± 0.03
0.083±0.002
0.062
0.88
1.45
0.012
Table 4.14. Linear regression parameters from reciprocal of rate of cellulose
degradation vs. reciprocal of Ca+2 concentrations, and μmax and Ks coefficients
derivated from the linear function of the reciprocals.
108
The Ks for FS-A3c was 0.204 mM Ca+2 and for FS-S85 was 0.012 mM Ca+2, and
the μMax was 2.01 mg cellulose/h for FS-A3c, and 1.45 mg cellulose/h for FSS85. These results indicate a tendency for higher affinity for Ca+2 by FS-S85 than
FS-A3c, thus confirming the apparently higher Ca+2 requirement of FS-A3c for
cellulose degradation.
4.5. Cellulose degradation by F. succinogenes with different Ca+2
concentrations in NH3-free media.
Because F. succinogenes was found to have an absolute requirement for
Ca+2 with cellulose as a substrate and because degradation also responded to
Ca+2 concentrations with considerable variation, a study was designed to
measure cellulose degradation in NH3-free media. This eliminates growth as a
confounding factor in the determination of Ca+2 requirement for cellulose
degradation.
The effects of strain, Ca+2 concentration and the interaction of
strain*treatment on cellulose degradation in NH3-free media were evaluated.
Maximum degradation was affected for all the effects tested (P<0.05), and rate of
degradation was significantly affected by Ca+2 concentration (P<0.05); the effect
of strain and the interaction of strain*treatment showed only a tendency (P= 0.13
and 0.08, respectively). Lag time was affected only by strain (P<0.05) and FSS85 showed longer lag times than FS-A3c.
For both strains of FS with the Ca+2-free medium, no cellulose degradation
occurred (Table 4.15); therefore there was no data to include in the overall
analysis. Very low degradation by FS-A3c was obtained with 0.02mM Ca+2, and
109
the data from this treatment fit neither the logistic nor linear model. Therefore
the least squares means were estimated for each sampling time and
comparisons among them were made to determine if degradation along time was
different or not. Differences were found between the two later sampling times 24
and 48 hours, and this difference was used to estimate the rate of degradation.
The LSM for 48 hours is reported as the extent of cellulose degradation: 3.9 mg.
Thus, cellulose degradation in both strains of F. succinogenes responded
to the Ca+2 concentrations following a definite pattern.
With 0 Ca+2 no
degradation occurred; as Ca+2 concentrations increase, cellulose degradation
also increased, both in extent and the rate of degradation (Table 4.15., Figure
4.26 and 4.27). FS-A3c showed a linear and quadratic response for maximum
cellulose degradation (P<0.0001), whereas FS-S85 has a linear (P=0.0029)
increasing in maximum degradation as Ca+2 concentration increased (Table 4.15
and Figure 4.28). Considering the significance of the contrasts and the curve of
maximum degradation vs. Ca+2 concentrations for A3c, the quadratic adjustment
is more likely to define the Ca+2 requirement for maximum degradation. Thus by
calculation of maximum from the quadratic function, Ca+2 requirement is 0.47
mM. It is not possible to calculate a requirement for FS-S85 because the linear
contrasts, although the curve in the figure 4.28 shows a tendency to be
quadratic. This may be due to the high variation of the measurements. Rate of
degradation responded to increased Ca+2 concentrations in a quadratic fashion
(Figure 4.29), statistically significant for FS-A3c (P=0.0126) and only as a
110
tendency for FS-S85 (P=0.0805). In this case, the requirement can be defined
as 0.4 mM for rate of degradation for FS-A3c. For both strains, lag time was not
affected by Ca+2 concentrations (P>0.05).
Max of
degradation**
Rate of
degradation***
Lag time****
F. succinogenes A3c
0*
0.02
0.04
0.08
0.16
0.32
0.64
No degradation
3.90 (at 48 h)
7.14 ± 1.22
11.23 ± 1.07
12.17 ± 1.21
11.32 ± 0.83
12.08 ± 1.15
0.11 (estimated)
0.30 ± 0.12
0.42 ± 0.09
0.41 ± 0.09
0.67 ± 0.09
0.50 ± 0.10
< 12 (estimated)
10.42 ± 3.45
9.95 ± 2.29
6.53 ± 2.41
9.15 ± 1.39
6.92 ± 2.40
Linear1
Quadratic
<0.0001
<0.0001
0.0560
0.0126
0.2030
0.6559
F. succinogenes S85
0*
0.02
0.04
0.08
0.16
0.32
0.64
No degradation
6.67 ± 1.36
8.73 ± 1.45
10.05 ± 0.11
11.26 ± 1.18
12.71 ± 0.59
11.03 ± 0.27
0.18 ± 0.08
0.39 ± 0.14
0.55 ± 0.06
0.53 ± 0.11
0.51 ± 0.05
0.80 ± 0.03
11.48 ± 6.88
11.07 ± 3.28
14.97 ± 0.16
8.62 ± 2.47
12.11 ± 1.10
15.95 ± 0.44
Linear1
Quadratic
0.0029
0.2737
0.0521
0.0805
0.9901
0.8366
* Ca+2 mM **mg of cellulose ***mg cellulose/hour ****hours
1
: for each strain, max degradation, rate of degradation and lag time were
evaluated by linear and quadratic contrasts, P values are presented.
Table 4.15. Cellulose degradation by Fibrobacter succinogenes strains A3c and
S85 incubated with different concentrations of Ca+2 and NH3-free media.
111
mg
14
12
0
0.02
10
0.04
0.08
8
0.16
0.32
6
0.64
4
2
0
0
20
40
60
80
hours
Figure 4.26. Predicted cellulose degradation curves of F. succinogenes A3c with
different Ca+2 concentrations in NH3-free cellulose media, using the parameters
obtained from the logistic function.
mg
14
12
0
10
0.02
0.04
8
0.08
0.16
6
0.32
0.64
4
2
0
0
20
40
60
80 hours
Figure 4. 27. Predicted cellulose degradation curves of F. succinogenes S85 with
different Ca+2 concentrations in NH3-free cellulose media, using the parameters
obtained from the logistic function.
112
mg
14
FS-A3c
12
FS-S85
10
8
6
4
2
Ca +2, mM
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Figure 4. 28. Maximum cellulose degradation vs. Ca+2 concentrations for F.
succinogenes strains A3c and S85, from NH3-free medium.
mg/h
0.9
FS-A3c
0.8
FS-S85
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Ca
0
0
Figure 4.29.
0.1
0.2
0.3
0.4
0.5
0.6
+2
, mM
0.7
Rate of cellulose degradation vs. Ca+2 concentrations for F.
succinogenes strains A3c and S85, from NH3-free medium.
113
Examining the data by the kinetic approach used previously, linear
functions of the reciprocals were obtained (Table 4.16. and Figure 4 24). μVmax
and Ks calculated were, respectively: 0.61 mg/h and 0.041 mM Ca+2 for FS-A3c,
and 0.68 mg/h and 0.056 mM Ca+2 for FS-S85.
Coefficients of linear function
P=
Strain
Intercept
FS-A3c*
FS-S85
1.66±0.22
1.47±0.15
r
2
μmax
Ks
(mg/h)
(mM)
Slope
0.068±0.02
0.029
0.84
0.61
0.041
0.08±0.006
0.0002 0.98
0.68
0.056
*without data from 0.02 mM
Table 4.16.
Linear regression parameters from reciprocal of rate of cellulose
degradation vs. reciprocal of Ca+2 concentrations, and μmax and Ks coefficients
derivated from the linear function.
Attachment is an important issue for cellulose degradation, and many
authors have studied this process with different experimental methods.
Akin
(1976, 1980) did morphological studies using the electron microscope, Minato
114
and Suto (1978, 1981) studied the mechanisms of how the bacterium attaches to
the fiber (cellulose binding proteins). Gong and Forsberg (1989) and Roger et al
(1990) evaluated the effect of different factors in this process among them,
different cations.
Gong and Forsberg (1989), using F. succinogenes S85, evaluated
different concentrations of Ca and Mg on cellulose adhesion.
They found that
the percentage of adhesion increased abruptly with 27.3 mM Ca or 20.2 mM Mg,
probably due to the ionic strength positively improving the bacterial adhesion,
rather than the effect of the cations per se. When the same authors evaluated the
effect of heat, trypsin, glutaraldehyde or pronase, bacterial adhesion to cellulose
was markedly reduced, suggesting that a protein and not cations was involved in
adhesion to cellulose of F. succinogenes.
Roger et al (1990) evaluated the lack of Ca or Mg, or both on adhesion of
R. flavefaciens (007) and F. succinogenes (S85) on cellulose. No changes in
adhesion for F. succinogenes were observed, and a reduced adhesion was
observed in R. flavefaciens when either cation was absent; the effect was greater
with the lack of Mg or Ca+Mg.
Concerning the interaction between different ions and plant cell walls,
Molloy and Richards (1971) reported that lignin and pectin can bind Ca+2 and
Mg+2, and Somers (1973) demonstrated that pectin in plant cell walls bind Ca+2.
Torre et al (1992) found that cellulose does not have the capacity to bind Ca+2,
while pectin has a large capacity to bind Ca+2. Consistent with this information;
results of cation exchange capacity (CEC) obtained by Allen et al (1985) showed
115
a very low CEC for Whatman paper or solka floc; and McBurney et al (1986)
found that CEC is correlated with lignin, lignin:ADF ratio, and nitrogen content,
but not with cellulose.
Considering the information from Gong and Forsberg (1989), Roger et al
(1990), who did not find a relation between the presence of Ca or Mg and
attachment of F. succinogenes to cellulose, and the findings from Molloy and
Richards (1971), Somers (1973), Allen et al (1985), McBuerney et al (1986) and
Torre et al (1992), who agreed that pectin and lignin, but not cellulose, are the
cell wall components responsible for the capacity to bind divalent cations; it is
possible to assume that Ca+2 did not have a direct effect on binding for F.
succinogenes to purified cellulose. The response in cellulose degradation must
be related either with the secretion of enzymes from the bacterial cell, or as an
intermediate in the attachment between the enzyme(s) and cellulose, or to the
amount and activity of the enzymes. Our experimental evidence is insufficient to
elucidate which of these mechanisms might be involved.
Numerous experiments have observed an increased in cellulose
degradation when Ca was added to the medium, either using whole bacteria or
isolated enzymes from cellulolytic bacteria.
Using washed rumen bacteria
Hubbert et al (1958a) suggested the optimal range for Ca supplementation was
0.34 – 2.04 mM.
With isolated endoglucanase-1 from F. succinogenes,
increasing the concentration from 1 to 10mM of CaCl2 (0.27 to 2.7mM Ca)
increased rate of enzyme activity, whereas endoglucanase-2 did not change its
activity (McGavin and Forsberg, 1988). Taylor et al (1987) using a Cel-encoded
116
endoglucanase expressed in E. coli, obtained an increased enzymatic activity
when Ca (10mM) was added to the media; however, no response was observed
using the native cellulolytic enzyme(s) from F. succinogenes. Tsai et al (2003)
characterizing the Fsβ-glucanase from F. succinogenes, observed that the
addition of 1 mM Ca+2 increased the thermal stability of the enzyme.
Phospholipase A (PLA) is an enzyme present in the outer membrane of
Gram (-) bacteria which is sensitive to Ca+2 for its activation.
This enzyme
causes changes in fatty acid composition, permeability and fluidity of the
membrane (Snijder and Dijkstra, 2000). It has been suggested that this is the
mechanism used by the bacteria to allow secretion of different substances
involved in pathogenicity of the bacteria (Snijder and Dijkstra, 2000). Christie
(1981) reported that PLA of Butyrivibrio fibrisolvens is not sensitive to Ca+2, but
no information about other rumen Gram (-) bacteria apparently is available.
Could this be a mechanism used for F. succinogenes to secret the enzymes
complex across the membrane?
4.6. Cellulose degradation by R. flavefaciens in NH3-free media, when the
medium is also Ca+2 and Mg-free, Ca+2-free, or Mg-free.
When growth experiments were analyzed, R. flavefaciens showed a high
max growth and rate of growth in Ca+2 free and low Ca+2 media.
absolute growth requirement of R. flavefaciens for Mg was observed.
Later an
When
cellulose degradation was studied, the increased Ca+2 concentration was
117
reflected in a small increment of cellulose degradation, characterized as a linear
response (P<0.0001). To separate the potentially confounding effect of growth
on cellulose degradation, due to Ca+2 or Mg, an experiment was carried out with
R. flavefaciens in NH3-free media. Treatments were: Ca+2 and Mg at normal
concentrations (control), Ca+2-free, Mg-free, and Ca+2 and Mg-free. The results
are shown in Table 4.17. For both strains the extent of degradation at 60 hours
incubation with Ca+2 and Mg was similar to Ca+2-free (P>0.05). Neither Mg-free
nor Ca+2 and Mg-free, showed any cellulose degradation. Thus when Mg was not
present in the NH3-free medium, no cellulose degradation occurred.
No
interaction between strain and cation treatment was observed (P>0.05). Strain
effect was observed (P<0.05), where RF-C94 showed lower cellulose
degradation.
This suggests that R. flavefaciens requires Mg to degrade
cellulose, and may has no requirement or requires very low concentrations of
Ca+2 for cellulose degradation; if not why no degradation occurred when the
media Mg-free+Ca+2 had 0.36 mM of Ca+2.
118
Cellulose media NH3-free
Strains
Ca+2+Mg
Ca+2-free
+ Mg
(Ca+2+Mg)-
Mg-free
free
+Ca+2
RF-B34b
14.38 ± 1.93 a
13.55 ± 1.93 a
ND
ND
RF-C94
10.35 ± 1.93 b
7.50 ± 1.93 b
ND
ND
mg of cellulose degraded
ND: no degradation
a, b: different letters indicate a strain effect on extent of degradation (P<0.05)
Table 4.17.
Extent of cellulose degradation (mg degraded) at 60 hours
incubation for Ruminococcus flavefaciens (B34b and C94) in the presence or
absence of Ca+2 and/or Mg in NH3-free cellulose liquid medium.
4.7. General discussion.
The results from the present study show different responses for Ca+2 and
Mg among species and strains of the predominant rumen cellulolytic bacteria:
Fibrobacter suucinogenes, Ruminococcus albus and Ruminococcus flavefaciens.
These three species are the predominant cultured cellulolytic bacteria found in
the rumen (Miron et al, 2001; Russell, 2002), and among these the mechanisms
for cellulose degradation have been investigated. Studies have been conducted
on nutrient requirements, type of enzymes produced, and mechanisms of
attachment to the substrate.
119
With the discovery of the cellulosome in Clostridium thermocelum by
Bayer and coworkers, in 1983, the enzymes and attachment to the substrate
were understood to be part of an integrated system secreted by the bacterium.
This provided a new approach to study the mechanisms that rumen bacteria can
use to degrade the insoluble carbohydrates. Since then, different groups of
researchers have advanced the characterization of the molecular structure from
different systems of attachment and substrate degradation. Also the genome
sequence has been characterizated for different species. Cellulosome, fimbria or
pili and glycocalix are the different systems found in rumen cellulolytic bacteria
(Miron et al, 2001).
Ruminoccocus albus has cellulosome and fimbria as
attachment systems, R. flavefaciens has cellulosome and glycocalyx, and F.
succinogenes apparently has only glycocalyx (Chesson and Forsberg, 1997;
Ohara et al, 2001, Rincon et al, 2001, Bayer et al, 2004).
Evidence from
molecular studies shows that many of the structures from cellulosome, fimbria,
and glycocalyx have proteins with peptide sequences and amino acid residues
which form an EF-hand motif that has the capacity to bind Ca+2, thus affecting
the molecular stability
or structure and perhaps also the activity of many
enzymes (Chavaux et al, 1990; Chavaux et al, 1995; Chio and Ljungdahl,
1996a,b; Kirby et al, 1997; Ohara et al, 2000; Mitsumori and Minato, 2000; Lyttle
et al, 2001, Rincon et al, 2001).
Due to the evidence that cellulosome, fimbria, and individual enzymes, all
involved in cellulose digestion, all requiring Ca+2; it is possible that the three
cellulolytic rumen bacterial species may have a specific requirement for this
120
divalent cation. There is no information related with Ca+2 requirements for either
R. albus or R. flavefaciens. For F. succinogenes, only total calcium requirement
for growth is available (Bryant et al, 1959), whereas, no information for Ca+2
requirement for cellulose degradation is available. Thus, this study explored the
Ca+2 requirements for the three predominant rumen cultured cellulolytic bacteria,
for growth as well as for cellulose degradation.
For maximum growth a statistically significant response was found with
increasing Ca+2 concentrations. FS-S85 responded in quadratic form, while RA-7
had a tendency for linear response. For FS-A3c, RF-B34b and RF-C94, the
contrasts analysis, to evaluated the form of the response, did not give a solution;
maybe due the special response from these bacteria to increased Ca+2
concentrations (Figure 4.8). Although, when rate of growth was evaluated, FSA3c showed a quadratic response to the increased Ca+2 concentrations (Figure
4.9).
An especial situation was observed with FS-A3c, in that under certain
conditions it did not grow in Ca+2-free medium. Further, when FS-A3c grew in
apparently Ca+2-free media, a small amount of Ca+2 was found in the medium,
apparently from contamination in the water supply. Thus a very small amount of
Ca+2 triggered a growth response in this bacterium. However, this growth was
not normal from the reproduction aspect, since FS-A3c showed an increased
pleomorphism under these conditions (≤ 0.02 mM Ca+2).
121
Although relatively normal growth was achieved with the very low Ca+2
concentrations, the Ca+2 requirement for maximum growth was higher than
expected. The biological importance of this response is may be questionable in
the majority of cases. The incremental response to increasing Ca+2 concentration
is small and in many cases highly variable (Figure 4.8). Relatively low Ca+2
concentrations which occur from contamination in reagent grade minerals, permit
normal growth, and for example, satisfy the growth of R. albus 7.
There is no logical explanation for the inconsistent results from this study
for FS-S85 compared with those obtained for the same strain by Bryant et al
(1959). In this study, FS-S85, reached maximum growth under all experimental
Ca+2 concentrations. Although FS-S85 had a quadratic response to increasing
Ca+2 concentrations, the incremental response was very small.
In contrast,
Bryant et al (1959) observed increasing growth response to increased Ca
concentrations and defined a requirement for this strain as 0.05 mg / 5 ml
(calculated as 0.25 mM total Ca), which is lower than this estimated from the
present results (0.47 mM as Ca+2).
The special response to increasing Ca+2 may be related to the function of
Ca+2 in bacterial cells. The evidence from aerobic pathogenic bacteria shows
Ca+2 has a role as a signaling molecule (Herbaud et al, 1998). Bacteria, similar
to eukaryotic cells, can regulate intracellular Ca+2 concentrations. They respond
to changes of Ca+2 concentrations in the external medium, depending on a
normal energy status, because the function of Ca+2 channels is energy
dependent (Rosen and Gangola, 1987). Changes in Ca+2 concentrations in the
122
external media are reflected by transient modification in intracellular Ca+2
concentrations; recovery to the steady state depends on the adaptation to
concentrations in the medium (Jones et al, 1999).
Thus if the media Ca+2
concentration is in the micromolar range they are able to internally accumulate 3
to 6 fold the Ca+2 concentration that is in the medium. However, if the medium
concentrations are in millimolar order, they tend to accumulate similar
concentrations as in the medium (Jones et al, 2002). Perhaps, F. succinogenes,
has regulatory mechanisms similar to E. coli, and can accumulated Ca+2 in the
cell.
Durand and Kawashima (1980), reported an increases in bacterial
pleomorphism when calcium was deficient in the culture medium. However, the
authors did not mention which species of bacteria were affected and gave no
explanation for this effect. Although, FS-A3c naturally has a tendency to have
some pleomorphic rods in older cultures (Dehority, 2003), this was not evident
when the bacterium was grown in normal medium.
When FS-A3c was grown
with low Ca+2 concentrations (0.02 mM Ca+2), pleomorphic forms were observed
and the bacteria grew in size but did not divide. Evidence to explain this effect
from Ca+2 deficiency came from studies with E.coli. It was found that Ca+2 plays
a regulatory role in cell division through FtsZ, a protein that require Ca+2 to
became polymerized, and is responsible to signal the cell equator and form the
division ring for cell division (Yu and Margolin, 1997). When Ca+2 is depleted in
the medium cells became elongated, forming filamentous multinucleated cells.
123
Unfortunately, in the present study no observations were made at other Ca+2
concentrations to determine what concentration is the threshold for this
phenomenon.
The lack of response from R. flavefaciens to increasing Ca+2
concentrations was not expected, based on the information reported by Durand
and Kawashima (1980), who suggested that these organisms have a higher
requirement for Ca and Mg because these elements are needed for synthesis of
the cell walls by Gram (+) bacteria. In general, Mg, as a cofactor, is involved in
many different steps in the synthesis of the cell walls (Ghuysen and Shockman,
1973), while no role has been identified in this process for Ca+2 , other than its
role in sporulation process (Silver, 1977). Both Ca+2 and Mg are part of the cell
wall structure, due to chemical interactions between compounds at the bacterial
cell wall and divalent cations in the medium. Thus, teichoic and/or teichuronic
acids present at the matrix of the peptidoglycan layer of Gram (+) bacteria can
bind Ca+2 and Mg from the medium and the amount bound will vary as a function
of the cell wall composition (Beveridge and Murray, 1980; Beveridge et al, 1982;
Beveridge, 1990).
It is necessary to consider that these cations, while they are
bound to the cell wall, are not available for exchange with the medium or other
cells. If the bacteria die or the pH conditions became acidic, these cations
become available in the medium (Durand and Kawashima, 1980). Thus, a kind
of mineral recycling can happen.
But in the present study, this option may not
apply, since the size of the inoculum utilized was very small (0.1 ml of 0.1 OD
culture).
124
Other structures of the cell envelope are the S-layers that are ubiquitous in
Gram (+) and Gram (-) bacteria. A role for Ca+2 in the assembly of the units of the
S-layer in Caulobacter crescentus and Campylobacter fetus fetus was described
by Sleytr and Beveridge (1999). Although no information is available for S-layers
in rumen bacteria, in theory they must be related with secretory processes,
possibly the cellulosome or extracellular enzymes.
For cellulose degradation there is without doubt an absolute Ca+2
requirement by F. succinogenes, because neither strain degraded cellulose at 0added Ca+2 concentration. Extent of degradation and lag time were affected by
the Ca+2 concentrations to different degrees. For maximum degradation, FS-A3c
responded linearly and FS-S85 did not reach solution for contrasts analysis. This
may be due to variation within strain, because the form of the curves described
for both are very similar (Figure 4,18).
When growth was excluded from the
process by deleting N from the medium, the results confirmed the specific
requirement of Ca+2 for cellulose degradation. Both extent and rate of
degradation were significantly affected by Ca+2 concentrations (Figure 4.28 and
4.29). The highest extent of degradation was attained with 0.42 mM, whereas
the highest rate of degradation was with 0.39 mM, respectively, for FS-A3c,
values that are very close. For FS-S85 a linear contrast was significant for both
maximum degradation and rate of degradation. When rate of degradation was
plotted vs Ca+2 concentrations (Figure 4.29), it was readily apparent that
maximum rate of degradation was attained when Ca+2 was around 0.32 and 0.64
mM of Ca+2, for FS-A3c and FS-S85, respectively, observation that is close to the
125
statistical estimation. The affinity (Ks) values for Ca+2 were estimated to be 0.041
and 0.056 mM Ca+2 for FS-A3c and FS-S85, respectively, which are very low
values compared with the concentrations at which what maximum rate of
degradation was obtained.
McGavin and Forsberg (1988) found that endoglucanase-1 from F.
succinogenes S85 increased in activity when CaCl2 was added to the medium (1
to 10 mM), whereas endoglucanase-2 did not respond; Mitsumori et al (2002)
reported that recombinant endoglucanase-F had an improved binding of the
enzyme to carboxymethylcellulose when Ca+2 was supplemented to the medium,
although no Ca+2 binding motif was identified in this enzyme. These inconclusive
data correspond to the varied responses observed for Ca+2 requirements in the
present studies.
As expected, cellulose degradation by R. albus and R. flavefaciens
increased with increased Ca+2 concentrations. Both species have cellulosome
structures which have amino acid sequences that bind Ca+2. Although RA-7
showed a response for maximum degradation, rate of degradation, and lag time,
RA-8 did not response to Ca+2 concentrations (P>0.05).
The very slow
degradation observed for that this bacterium may have affected its response. All
factors known to be needed for R.albus 8 to degrade cellulose were included in
the medium (Stack et al, 1983).
126
F. succinogenes has a very unstable bacterial membrane that apparently
permits the secretion of the enzyme as blebs on the cell envelope (Huang and
Forsberg, 1988); this could be related to the chemical structure of the cell
envelope, as well to the presence of phospholipases A and C on the bacterial
wall, but there is no available information to confirm it.
As mention earlier, E.
coli, B. cereus and C. welchii, have phospholipase A and C, which are activated
by Ca+2, modifying its lipid composition, and the membrane become more fluid;
permiting the secretion of proteins involved in their pathogenic capacity (Silver,
1977). Perhaps similar mechanisms can occur with F. succinogenes to secrete
their enzymes; since was observed a positive response to Ca+2 for cellulose
degradation.
Succinic acid, a normal metabolic end product for F. succinogenes
(Russell, 2002, Dehority, 2003), can chelate Ca+2 to form calcium succinate, a
substance that is slightly soluble in water (Windholz, 1976).
Under rumen
conditions, succinic acid is rapidly used by other bacteria and is not freely
available. However, in this study working with individual strains, F. succinogenes
could have produced succinic acid that would have accumulated and formed
calcium salts, thus chelating part of the available Ca+2. Because of the neutral
pH of the medium and the pKa of succinic acid (2.5), chelation would be strongly
in the salt form. However, if this happens, it occurs after the energy source
becomes available (cellobiose or cellulose).
The most likely situation for this to
occur would have been when bacteria were grown on cellobiose, a soluble
carbohydrate; but in general the response to Ca+2 observed with F. succinogenes
127
growing with cellobiose, suggest that if succinic acid was chelating Ca+2 it would
have resulted in decreasing Ca+2 concentration that did not affect the growth of
the bacterium.
The present results are supported by the evidence available in the
literature. It is reported that Ca+2 binding improves the stability of the structures
involved, or permits the interaction between dockerin domains and catalytic
domains, or improves enzymatic activity, as was demonstrated for C.
thermocellum (Chavaux et al, 1990 and 1995; Choi and Ljungdahl,1996, 1996b;
Bayer et al, 1998; Shoham et al, 1999; Lytle et al, 2001). The evidence for R.
flavefaciens came studies from Kirby et al (1997), Aurilia et al, (2000), Ding et al
(2001), and Rincon et al (2001); and for R. albus, from Ohara et al, (2000).
Also, the information available, from washed rumen bacteria in response
to Ca for cellulose degradation, is supporting the positive results obtained at the
present study. Although, the concentrations used for different authors are high
and very variable among studies; Hubbert et al (1958a) suggest as optimal
concentration a wide range of 50 - 300 ug/ml of CaCl2, (0.34 to 2.04 mM total Ca)
with a toxic level of 450 ug/ml of CaCl2 (11.23 mM total Ca).
Uesaka et al
(1967), cited by Durand and Kawashima (1980), reported a positive response by
increasing 10 to 100 ug/ml of CaCl2 (0.068 to 0.68 mM total Ca) and Bales et al
(1978) reporting an increased DM digestion of corn stalk when Ca increased
from 10 to 190 ppm (0.068 to 1.29 mM total Ca).
128
If the response from the present experiments is extrapolated to the
general rumen ecosystem, considering the Ca+2 concentrations reported by
Palmquist et al (1986) and Wagner et al (1993), who are the only authors that
measured ionized calcium in rumen fluid, it is not possible to envision a
deficiency. Reported Ca+2 concentrations were 0.6 mM at 0 h, and 1.5 mM at 10
h (Palmquist et al, 1986); and 1.75, 7, and 0.25 mM for 3, 6, and 12 h post
prandial (Wagner et al, 1993); all higher than the apparent requirements
observed in this study.
Fluctuations in Ca+2 concentration observed as a
consequence of the feeding time apparently have no consequences on bacterial
activity, although the Ca+2 released from different forages or diets and the
solubility of the supplemental sources could have different time patterns after
feeding.
When no response to Ca+2 for growth was observed with R. flavefaciens,
possible divalent cation requirements for growth were investigated. Magnesium
was the only cation identified as an absolute requirement and only for R.
flavefaciens. Growth requirements for Mg were then measured in all species,
and all responded to increased Mg concentrations, either in maximum growth
(Figure 4.16), rate of growth (Figure 4.17) or lag time.
highest requirement.
R. flavefaciens has
FS-A3c did not fit with linear or quadratic response,
although Mg concentrations affected the growth of the strain (Table 4.9, and
Figure 4.16). The results obtained for Mg requirement for growth with FS-S85
are higher than the results obtained by Bryant et al (1959) for the same strain
(0.56 mM vs. 0.1 mM Mg).
129
The high requirements for Mg by R. flavefaciens and R. albus are
somewhat logical since both are Gram (+) bacteria and they have higher Mg
requirements for cell wall synthesis (Ghuysen and Shockman, 1973). The higher
requirements by R. flavefaciens than R. albus may be due to differences in the
composition of the cell wall (Beveridge and Murray, 1980; Beveridge et al, 1982;
Beveridge, 1990), or to the mechanisms by which Mg is interchanged between
the intracellular and extracellular environments (Moncrief and Maguire et al,
1999; Chamnongpol and Groisman, 2002). Unfortunately, no specific information
about this subject is available for rumen bacteria.
Because R. flavefaciens C94 tended to have a high max growth and rate
of growth at 0 mM Ca+2, and had a small, linear, response to increased Ca+2
concentration in cellulose degradation; and also has a high Mg requirement for
growth; NH3-free medium was used to evaluate a possible role of Mg for R.
flavefaciens in cellulose degradation. Using a large inoculum of bacteria it was
observed that RF-B34b and RF-C94 did not degrade cellulose when Mg was
deleted from the medium.
No effect on cellulose degradation was observed
when Ca+2 was omitted.
The summary of results for Ca+2 and Mg requirements for cellulolytic
rumen bacteria are presented in table 4.18. The criteria used were maximum
growth and extent of cellulose degradation.
130
For growth and Ca+2, FS-S85 and RA-7 appear to have 0.47 mM and
>0.64 mM Ca+2 as requirement for growth, respectively. While no requirement
estimations for FS-A3c, RF-B34b and RF-C94 were made, because the lack of
solution on contrasts analysis. RA-8 did not respond to Ca+2 concentrations on
maximum growth, and no requirements can be establish for this cation.
For Mg, only R. flavefaciens has an absolute requirement for growth. For
FS-A3c no estimation can be made. Because RF-C94 responded linearly to Mg,
the requirement may be higher than 0.82 mM Mg. For FS-S85, RA-7, RA-8, and
RF-B34b Mg requirements were estimated trough max growth as 0.56, 0.52, 0.51
and 0.54 mM, respectively.
For cellulose degradation only F. succinogenes has an absolute Ca+2
requirement, either in NH3-normal or NH3-free medium; requirements for FS-A3c
are higher than 0.36 mM of Ca+2 in NH3-normal cellulose medium and 0.42 mM
of Ca+2 in NH3-free cellulose medium. Despite the absolute Ca+2 requirement for
cellulose degradation for FS-S85 no estimation can be made from NH3-normal
cellulose medium. FS-S85 in NH3-free medium showed a linear response and
then requirement may be higher than 0.64 mM Ca+2. When NH3-free medium
was used, and rate of degradation was the criterium, with 0.39 and 0.44 mM
Ca+2 maximum rate of degradation was attained, for FS-A3c and FS-S85,
respectively (Tables 4.11 and 4.15).
For cellulose degradation, R. albus 7, and R. flavefaciens, B34b and C94,
responded to increased Ca+2 concentrations, apparently RA-7 reach the
maximum degradation at 0.28 mM; and RF-B34b and RF-C94 had a significant
131
linear response, although this can be a mathematical adjustment to the more
complex behavior as is evident in the chart, and may not represent the biology of
the response (Figure 4.24).
Maximum growth
Cellobiose
1
Maximum degradation
Cellulose
Ca+2
1
F. succinogenes A3c
n.e.
n.e.
>0.36*
0.42*
F. succinogenes S85
0.47
0.56
n.e.*
>0.64*
R. albus 7
>0.64**
0.52
0.28
-
R. albus 8
n.r.
0.51
n.r.
-
R. flavefaciens B34b
n.e.
0.54*
>0.64
-
R. flavefaciens C94
n.e.
>0.82*
>0.64
-
Mg
1
Ca+2
Cellulose
NH3-free
1
Ca+2
1
Ca+2 and Mg concentrations as mM.
* For the bacterium and respective substrate there is an absolute cation
requirement.
** statistical tendency, (0.05<P>0.10).
n.e. no estimable, because no solution for contrasts.
n.r. no requirement can be established, because no response to the specific
cation concentrations.
Table 4. 18.
Summary of Ca+2 and Mg requirements for cellulolytic rumen
bacteria using maximum growth and extent of cellulose degradation as criteria;
Ca+2 requirements for growth and cellulose degradation, and Mg requirements
for growth.
132
I have concerns related with the response in cellulose degradation and
possible Ca+2 requirements for FS-S85, RF-B34b, and RF-C94. FS-S85 look
very alike with FS-A3c response, but the lack of solution from the statistical
analysis limit the estimation under these conditions. The linear response of R.
flavefaciens strains does not appear so clear, as was mention before.
The
possible the role that Mg has in cellulose degradation for R. flavefaciens need to
be consider. Because when Mg was absent of NH3-free cellulose media no
degradation occurs, despite the normal concentration of Ca+2.
May the
variations in the response to increased Ca+2 concentrations was the reflection of
Mg content in those media, and not the direct effect of Ca+2 concentrations. Also
a possible interaction between the two cations need to be evaluate. These are
aspects that deserve more attention and research.
The determination of mineral requirement for bacteria and their different
functions, may appear simple; however, the differences among species and
strains and the lack of information about their basic physiological functions in
rumen bacteria make this area of research somewhat complex. Without doubt,
more research needs to be undertaken, considering the variety of responses
obtained.
By using a higher number of concentration levels, increasing the
number of replicates, using larger size of culture, (particularly for cellulose
degradation, so multiple samples can be taken from the same vessel, to
decrease the error for been inoculating so many individual tubes) it should be
possible to achieve more precision in the data.
133
This should allow better
estimates of requirements and reflect the biology of the bacteria. In this study,
the statistic methods used to determine requirements have not always been
satisfactory to reach a solution; possibly due to high variation and also because
the particular response of the bacteria to the Ca+2 or Mg concentrations, which
no necessarily describes a linear or quadratic function.
In light of the results obtained in this study, there are many questions that
need to be addressed in future research. The role of Ca+2 and Mg in the rumen
bacterial cell wall (ion channels, enzymes, receptors, lipid composition, etc.).
Identify what cellulolytic enzymes are affected (amount, activity) by Ca+2 or other
cations such as Mg.
Identify why Ca+2 is an absolute requirement for F.
succinogenes in cellulose degradation. Is Ca+2 having a signalling role for cell
division in F. succinogenes, as it has in E. coli?. Why Mg is required in absolute
form for growth and apparently, also, for cellulose degradation for R.
flavefaciens?.
134
CONCLUSIONS
From the present study it is concluded that:
1. The different rumen cellulolytic species studied responded differently to
Ca+2 and Mg supplementation in the growth medium.
2. No Ca+2 requirement could be demonstrated for growth for Ruminoccocus
flavefaciens.
3. Although relative normal growth was achieved with very low Ca+2
concentrations, the Ca+2 requirement to maximize growth was higher than
expected.
The biological importance of the concentration required for
maximum response is questionable in the majority of the cases.
Relatively low concentrations can permit normal growth, as the small
amount of Ca+2 contained by the reagent grade minerals used to prepare
the media.
135
4. Fibrobacter succinogenes A3c showed some evidence of a special Ca+2
requirement. Pleomorphism was observed at lower Ca+2 concentrations,
suggesting that Ca+2 acts as a signal for cell reproduction.
5. Only Fibrobacter succinogenes, both strains, has an absolute Ca+2
requirement for cellulose degradation.
6. Ca+2 requirement for cellulose degradation for Ruminococcus flavefaciens
and R. albus was demonstrated, athough is not an absolute requirement.
7. Only Ruminococcus flavefaciens has an absolute Mg requirement for
growth.
8. For growth all the species and strains studied responded to increasing Mg
concentrations, having an specific requirement.
9. Ruminococcus flavefaciens also exhibits a Mg requirement for cellulose
degradation.
136
APPENDIX
137
APPENDIX A
Stock solutions, and different anaerobic media.
1.- Stock solutions:
12 % Sodium Carbonate:
Thirty-six g of anhydrous Na2CO3 are dissolved in deionized H2O and brought to
a final volume of 300 ml. The solution is transferred to a 500 ml round-bottom
flask and bubbled with N2 for at least 30 minutes. Tube in 10 ml aliquots under a
stream of N2. Sterilize the tubes at 120° C° for 20 min.
3% Cysteine⋅HCl:
Add 300 ml deionized H2O to 500 ml round-bottom flask. Heat to dispel gasses
while gassing with N2. When fairly well reduced (about 20-30 minutes), add 9.0 g
cysteine⋅HCl. Continue gassing with N2 for 15 minutes. Tube in 5.0 and 10.0 ml
aliquots under N2. Autoclave in tubes at 120° C° for 20 minutes.
0.1% Resazurin solution:
Add 0.1 g of resazurin to 100 ml volumetric flask and make to volume with
deionized water.
0.1% Hemin solution:
100 mg of hemin is dissolved in 0.02% NaOH and then diluted to a final volume
of 100 ml with deionized water.
138
Vitamin mixture: (1 liter) (Scott and Dehority, 1965)
200 mg Pyridoxine HCl
200 mg Riboflavin
200 mg Thiamine HCl
200 mg Nicotinamide
200 mg Ca-D-Pantothenate
10 mg Paraaminobenzoic acid (PABA)
5 mg Folic acid
5 mg Biotin
0.5 mg B12 Vitamin,
make to 1 liter volume with deionized water.
Volatile fatty acid mix: (31 ml) (Caldwell and Bryant, 1966)
17 ml Acetic acid (54.84 % v/v)
6 ml Propionic acid (18.75%)
4 ml Butyric acid (12.9%)
1 ml Isobutyric acid (3.23%)
1ml n-Valeric acid (3.23%)
1 ml Isovaleric acid (3.23%)
1 ml 2-methyl Butyric acid (3.23%)
Mineral solution I: (Bryant and Burkey, 1953)
0.3 % K2HPO4
Mineral solution II: (Bryant and Burkey, 1953)
0.3 % KH2PO4
0.6 % (NH4)2SO4
0.6 % NaCl
0.06 % MgSO4
0.06 % CaCl2⋅2H2O
139
Mineral solution A: (Scott and Dehority, 1965)
4.5 g/l KH2PO4
Mineral solution B: (Scott and Dehority, 1965)
4.5 g/l NaCl
4.5 g/l (NH4)2SO4
0.33 g/l CaCl2⋅2H2O
0.25 g/l MgSO4
0.10 g/l MnSO4⋅H2O
0.10 g/l FeSO4⋅7H2O
0.10 g/l ZnSO4⋅7H2O
0.01 g/l CoCl2⋅6H2O
Mineral solution B, NH3-free: (modified from Scott and Dehority, 1965)
4.5 g/l NaCl
0.33 g/l CaCl2⋅2H2O
0.25 g/l MgSO4
0.10 g/l MnSO4⋅H2O
0.10 g/l FeSO4⋅7H2O
0.10 g/l ZnSO4⋅7H2O
0.01 g/l CoCl2⋅6H2O
Mineral solution divalent-cation-free: (modified from Scott and Dehority,
1965)
4.5 g/l NaCl
4.5 g/l (NH4)2SO4
3.0 g/l KH2PO4
140
Mineral solution B, selective cation-free: (modified from Scott and Dehority,
1965)
To achieve selective cation-free media, the cation to be eliminated was not
included in Mineral solution B. Thus Mineral solution B, Ca+2-free, was prepared
by eliminating CaCl2; Mineral solution B, Mg-free was prepared by eliminating
MgSO4; Mineral solution B, Ca+2-Mg-free was prepared by eliminating both CaCl2
and MgSO4.
Phosphate buffer:
10 ml 7.5% NaH2PO4
10 ml 7.5% K2HPO4
280 ml of deionized H2O
Tube 9 ml aliquots without gassing, stopper with rubber stoppers and autoclave
at 120° C° for 20 minutes.
2.- Media:
Rumen fluid-glucose-cellobiose-agar slants medium (RGCA): (600 ml)
90 ml Mineral solution I
90 ml Mineral solution II
0.6 ml of 0.1% Resazurin solution
1.2 g glucose
1.2 g cellobiose
9.0 g agar
150 ml distilled water
240 ml rumen fluid (Strained through a double layer of cheesecloth and
centrifuged 10 minutes at 1000 x g).
Gas with CO2, bring almost to a boil over Bunsen burner, seal anaerobically, and
autoclave for 5 minutes at 100° C°. Place flask in 45° C° water bath, open
anaerobically under CO2 and add 20 ml of 12% Na2CO3 and 10 ml of 3%
141
cysteine HCl, continue gassing, check and adjust pH to 6.7-6.9 if needed with
NaOH or HCl. Tube 7.0 ml aliquots anaerobically into 16 x 150 mm culture tubes
from slants. Autoclave at 120° C° for 20 minutes in racks.
Complete liquid medium (cellobiose or cellulose substrate) with varing
levels of Ca+2 or Mg:
To prepare 300 ml:
60 ml mineral solution A
60 ml mineral solution B * / **
0.3 ml 0.1% rezasurin
0.3 ml 0.1% hemin
3 ml vitamin mix
1.5 g cellobiose (or 75 ml 3% cellulose ball-milled, Sigmacell® Type 20)
0.5 g trypticase
1.35 ml VFA mix
155 ml (or 85 ml) of deionized water
Adjust to pH 6.7- 6.9 with NaOH/HCl (1N)
Bubble with O2-free CO2, to dispel gasses; add 10 ml 12% Sodium carbonate
and continue gassing with CO2 for 20 to 30 minutes. Add 10 ml 3% cysteine and
keep gassing with CO2. Add CaCl2 or MgSO4, according to the experimental
medium and concentrations desired, continue gassing until reduced. Check and
adjust to pH 6.7-6.9 with NaOH or HCl. Tube 7 ml under CO2. Sterilize at 120°
C° by 20 minutes.
* To prepare media with different Ca+2 concentrations, Ca+2-free Mineral solution
B was used, and to achieve the different Ca+2 concentrations different amounts
of CaCl2 were added to the medium.
142
** To prepare media with different Mg concentrations, Mg-free Mineral solution B
was used, and to achieve the different Mg concentrations different amounts of
MgSO4 were added to the medium.
Due to the special requirement for R. albus 8, Phenylacetic acid (PAA)
and Phenypropionic acid (PPA) were added to the medium (Stack et al, 1983), to
achieve 0.25 uM as final concentration (Reveneau, 2001).
Divalent-Cation-free cellobiose liquid media:
To prepare 300 ml:
50 ml mineral solution divalent cation free
0.3 ml 0.1% rezasurin
0.3 ml 0.1% hemin
3 ml vitamin mix
1.5 g cellobiose
0.5 g trypticase
1.35 ml VFA mix
225 ml of deionized water
Adjust to pH 6.7- 6.9 with NaOH/HCl (1N)
Bubble with O2-free CO2, dispel gasses add 10 ml 12% Sodium carbonate and
continue gassing with CO2 for 20 to 30 minutes. Add 10 ml 3% cysteine and
continue gassing with CO2. Check and adjust to pH 6.7-6.9 with NaOH or HCl.
Tube 7 ml under CO2. Sterilize at 120° C° by 20 minutes.
143
NH3-free media cellulose liquid media
To prepare this media, NH4SO4, hemin and trypticase were eliminated, and
cysteine was reduced to half of normal concentration.
To prepare 300 ml:
60 ml mineral solution A
60 ml mineral solution B without (NH4)2SO4
0.3 ml 0.1% rezasurin
3 ml vitamin mix
75 ml 3% cellulose ball-milled
1.35 ml VFA mix
90 ml of deionized water
Adjust to pH 6.7- 6.9 with NaOH/HCl (1N)
Bubble with O2-free CO2, to dispel gasses. Add 10 ml 12% Sodium carbonate
and continue gassing with CO2 for 20 to 30 minutes. Add 5 ml 3% cysteine and
keep gassing with CO2. Check and adjust to pH 6.7-6.9 with NaOH or HCl.
Tube 7 ml under CO2. Sterilize at 120° C° by 20 minutes.
For NH3-free with different Ca+2, different amounts of 1M CaCl2 were added to
obtain the experimental concentrations.
For NH3-free Ca+2-free / Mg-free / Ca+2-Mg-free, specific mineral solution B NH3free were prepared, deleting the respective cation / cations.
144
APPENDIX B
Effect of consecutive transfer on Ca+2-free, Mg-free, Ca+2-Mg-free and
Normal media on growth (Max OD) by the predominant rumen cellulolytic
bacteria.
145
Liquid Growth Medium (cellobiose)1
Normal
media
Ca2+-free
Mg-free
Ca2+- Mg-free
1st transfer
1.04 (25h)2
0.70 (25h)
n.g.
1.35 (20h)
2nd transfer
1.00 (60h)
0.80 (36h)
n.g.3
0.70 (20h)
n.g.
1.4 (15h)
1.1(12h)
F. succinogenes S85
F. succinogenes A3c
1st transfer
2nd transfer
0.66 (16h)
3rd transfer
0.66 (16h)
R. albus 7
1st transfer
1.19 (14h)
0.52 (18h)
0.60 (25h)
2nd transfer
1.06 (11h)
0.53 (17h)
0.56 (40h)
3rd transfer
1.1(12h)
0.45 (22h)
R. albus 8
1st transfer
0.14 (167h)
0.56 (129h)
n.g.
n.g.
0.88(17h)
n.g.
n.g.
0.74 (28h)
0.23 (142h) 0.43 (170h)
R. flavefaciens B34b
1st transfer
0.90 (20h)
2nd transfer
0.86 (17h)
3rd transfer
0.87 (19h)
R. flavefaciens C94
1st transfer
0.92 (27h)
2nd transfer
0.95 (29h)
3rd transfer
0.45 (24h)
1
Calcium concentration in Ca+2-free medium: 0.046 and 0.014 mM as total and
ionized, respectively.
Magnesium concentration in Mg-free medium: 0.3 ug/ml total Mg.
Normal medium: 0.36 mM Ca+2 and 0.41 mM Mg
2
3
Maximum Od and hours to reach max OD
n.g. no growth.
146
APPENDIX C
Quadratic functions used to estimate the maximum point of the curve.
1. Ca+2 Growth experiments.- Maximum growth (A) (OD units)
FS-S85
A=1.2404 + 0.5424x –0.05805x2
(P=0.0205; r2=0.27).
- Rate of growth (B) (OD units/hour)
FS-A3c
B= 0.1468 + 0.2745x – 0.3216x2
(P=0.0001; r2=0.72)
RA-7
B= 0.2607 – 0.1375x + 0.1905x2
(P=0.2131; r2=0.12)
2. Magnesium Growth experiments.(Equations calculated with Magnesium concentration in mg/ml; and results are
reported in tables and figures as mM)
-Maximum growth (A) (OD units)
FS-S85
A= 0.9851 + 65.1899x – 2385.65x2
(P=0.0001; r2=0.72 )
RA-7
A= 0.9296 + 58.1543x – 2287.38x2
(P=0.0073; r2=0.37)
RA-8
A= 0.4861 + 17.3054x – 704.8414x2 (P=0.0203; r2=0.31)
RF-B34b
A= 0.4566 + 71.1056x – 2715.37x2
(P=0.0001; r2= 0.78)
-Rate of growth (B) (OD units/hour)
RF-B34b
B= 0.0478 + 7.48237x – 2777.69x2 (P=0.0258; r2=0.35)
147
-Lag time (C) (hours)
RF-B34b
C= 14.4296 - 572.90X + 16220x2
(P=0.0021; r2=0.52)
RF-C94
C= 22.4134 – 2165.48x + 81667x2 (P=0.0004; r2=0.60)
3. Ca+2 Cellulose degradation experiments
(Equations calculated as g, in tables and figures are expressed as mg)
-Maximum degradation (A) (g)
RA-7
A= 0.03452 + 0.01497x – 0.02661x2
(P=0.1440; r2=0.12)
-Lag time (C) (hours)
RA-7
C= 14.19354 – 35.8740x + 53.5384x2
(P<0.0001; r2 =0.54)
RF-C94
C= 17.5969 – 2.80475x - 14.0903x2
(P=0.2659; r2=0.06)
4. Ca+2 Cellulose degradation NH3-free experiments
(Equations calculated as g, in tables and figures are expressed as mg)
-Maximum degradation (A) (g)
FS-A3c
A= 0.00869 + 0.01549x – 0.01644x2
(P=0.0328; r2=0.35)
-Rate of degradation (B) (g/hour)
FS-A3c
B=0.00022248 + 0.00209x – 0.00258x2 (P=0.0106; r2=0.43)
148
BIBLIOGRAPHY
1.
Allen, M.S., M.I. McBurney, P.J. Van Soest. 1985. Cation-exchange
capacity of plant cell walls at neutral pH. J. Sci. Food Agric. 36:1065-1072.
2.
Aurilia, V., J. Martin, S. McCrae, K. Scott, M.T. Rincon, H. Flint. 2000.
Three multidomain esterases from the celulolytic rumen anaerobe
Ruminococcus flavefaciens 17 that carry divergent dockerin sequences.
Microbiology 146:1391-1397.
3.
Bales, G.L., D.W. Kellogg, N.S. Urquhart. 1978. Effects of certain
inorganic elements and urea on in vitro dry matter disappearance of milo
stalks. J. Anim. Sci. 47: 561-568.
4.
Barth, J., S.L. Hansard. 1962. Comparative availability of phytin and
inorganic phosphorus to rumen microorganisms, in vitro. Proc. Soc. Expt.
Biol. Med. 109:448-451
5.
Bath, S., E. Owen, M. Bhat. 2001. Isolation and characterization of major
cellobiohydrolase (S8) and a major endoglucanase (S11) subunit from the
cellulosome of Clostridium thermocellum. Anaerobe 7:171-179.
6.
Bayer, E.A., E. Morag, R. Lamed. 1994. The cellulosome – a treasuretrove for biotechnology. Trends Biotechnology 12:378-386.
7.
Bayer, E.A., H. Chanzy, R. Lamed, Y. Shoham. 1998. Cellulose,
cellulases and cellulosomes. Current Opinion in Structural Biology 8:548557.
149
8.
Bayer, E.A., J.P. Belaich, Y. Shoham, R. Lamed. 2004. The
cellulososmes: multienzymes machines for degradation of plant cell wall
polysaccharides. Annu. Rev. Microbiol. 58:521-554.
9.
Bennink, M.R., T.R. Tyler, G.M. Ward, D.E. Johnson. 1978. Ionic milieu of
bovine and ovine rumen as affected by diet. J. Dairy Sci. 61: 315-323.
10.
Beveridge, T.J., C.W. Forsberg, R.J. Doyle. 1982. Major sites of metal
binding in Bacillus licheniformis walls. J. Bacteriol. 150:1438-1448.
11.
Beveridge, T.J., R.G.E. Murray. 1980. Sites of metal deposition in the cell
wall of Bacillus subtilis. J Bacteriol. 141:876-887.
12.
Beveridge, T.J., S.F. Koval. 1981. Binding of metal to cell envelopes of
Escherichia coli K-12. Appl. Environ. Microbiol. 42:325-335.
13.
Bryant, M.P. 1973. Nutritional requirements of the predominant rumen
cellulolytic bacteria. Fed. Proc. 32:1809-1013.
14.
Bryant M.P., and L.A. Burkey. 1953. Cultural methods and some
characteristics of some of the more numerous groups of bacteria in the
bovine rumen. J. Dairy Sci. 36:205-217.
15.
Bryant, M.P. and R.N. Doetsch. 1954. A study of actively cellulolytic rodshaped bacteria of the bovine rumen. J. Dairy Sci. 37:1176-1183.
16.
Bryant, M.P., N. Small, C. Bouma, and I.M. Robinson. 1958.
Characteristics of ruminal anaerobic cellulolytic cocci and Cillobacterium
cellulosolvens n. sp. J. Bacteriol. 76:529-537.
17.
Bryant, M.P., I.M. Robinson, H. Chu. 1959. Observations on the nutrition
of Bacteroides succinogenes- a ruminal cellulolytic bacterium. J. Dairy Sci.
42:1831-1847.
150
18.
Caldwell, D.R., and M.P. Bryant. 1966. Medium without rumen fluid for
non-selective enumeration and isolation of rumen bacteria. Appl. Environ.
Microbiol. 14:1375-1381.
19.
Caldwell, D. R. and C. Arcand. 1974. Inorganic and metal-organic growth
requirements of the genus Bacteroides. J. Bacteriol. 120:322-333.
20.
Caldwell, D.R. and R.F. Hudson. 1974. Sodium, an obligate growth
requirement for predominant rumen bacteria. Appl. Microbiol. 27:549-552.
21.
Caldwell, D.R., M. Keeney, J.S. Barton, J.F. Keller. 1973. Sodium and
other inorganic growth requirements of Bacteroides amylophilus. J.
Bacteriol. 114:782-789.
22.
Chamnongpol, S., E.A. Groisman. 2002. Mg2+ homeostasis and
avoidance of metal toxicity. Mol. Microbiol. 44:561-571.
23.
Chavaux, S., P. Beguin, J.P. Aubert, K.M. Bhat, L.A. Gow, T.M. Wood, A.
Bairoch. 1990. Calcium-binding affinity and calcium-enhanced activity of
Clostridium thermocellum endoglucanase D. Biochem. J. 265:261-265.
24.
Chavaux, S., H. Souchon, P. Alzari, P. Chariot, P. Beguin. 1995.
Structural and functional analysis of the metal-binding sites of Clostridium
thermocellum endoglucanase CelD. J. Biol. Chem. 270:9757-9762.
25.
Chesson,A., C.W. Forsberg. 1997. Polysaccharide degradation by rumen
microorganisms. In: The rumen microbial ecosystem. Ed. Hobson, P.N.,
Stewart, C.S. 2nd ed. Blackie Academic and Professional. London, UK.
329-381 pp.
26.
Choi, S., L. Ljungdahl. 1996a. Dissociation of the cellulosome of
Clostridium thermocellum in the presence of ethylendiaminetetraacetic
acid occurs with the formation of truncated polypeptides. Biochemistry
35:4897-4905.
151
27.
Choi, S., L. Ljungdahl. 1996b. Structural role of calcium for the
organization of the cellulosome of Clostridium thermocellum. Biochemistry
35:4906-4910.
28.
Dehority, B.A. 1963. Isolation and characterization of several cellulolytic
bacteria from in vitro rumen fermentations. J. Dairy Sci. 46:217-222.
29.
Dehority, B.A. 1969. Pectin-fermenting bacteria isolated from the bovine
rumen. J. Bacteriol. 99:189-196.
30.
Dehority, B.A. 2003. Chapter 8. Cellulose digesting rumen bacteria. IN:
Rumen Microbiology. Nottingham. University Press, UK. Pp 177-208.
31.
Devendra, C.; D. Lewis. 1974. The interaction between dietary lipids and
fibre in the sheep. Anim. Prod. 19:67.
32.
Ding, S.Y., M.T Rincon, R. Lamed, J. Martin, S. McRae, V. Aurilia, Y.
Shoham, E. Bayer, H. Flint. 2001. Cellulosomal scaffoldin-like proteins
from Ruminococcus flavefaciens. J. Bacteriol. 183:1945-1953.
33.
Durand, M. R. Kawashima. 1980. Influence of minerals in rumen microbial
digestion. In: Digestive physiology and metabolism in ruminants.
Ruckebusch, Y. and P. Thivend Editors. MTP Press Limited. UK. 375 –
408 pp.
34.
El-Hag, G.A., T.B. Miller. 1972. Evaluation of whisky distillery by-products.
J. Sci. Food Agr. 23:247.
35.
Emanuele, S.M., C.R. Staples. 1990. Ruminal release of minerals from six
forage species. J. Anim. Sci. 68:2052-2060.
36.
Ferlay, A., M. Doreau. 1995. Influence of method of administration of
rapeseed oil in dairy cows. 2. Status of divalent cations. J. Dairy Sci.
78:2239-2246.
152
37.
Fitt, T.J.; K. Hutton, W.R. Otto. 1974. Relative affinities of isolated cell
walls of rumen bacteria for calcium and magnesium ions. Proc. Nutr. Soc.
333:106A-107A. (Abstract).
38.
Fitt, T.J., K. Hutton. 1974. Effect of potassium ions on the uptake of
magnesium by isolated cell walls of rumen bacteria. Proc. Nutr. Soc.
333:107A-108A. (Abstract).
39.
Galbraith, H., T.B. Miller, A.M. Paton, J.K. Thompson. 1971. Antibacterial
activity of long chain fatty acids and the reversal with Calcium,
Magnesium, Ergocalciferol and Cholesterol. J. Appl. Bact. 34:803-813.
40.
Gangola P., B.P. Rosen. 1987. Maintenance of intracellular Calcium in
Escherichia coli. J. Biol. Chem. 262:12570-12574.
41.
Ghuysen, J.M. and G.D. Shockman. 1973. Chapter 2: Biosynthesis of
peptidoglycan. In: Bacterial membranes and walls, Ed by L. Leive, M.
Dekker, Inc. NY. pp: 39-130.
42.
Grace, N.D., E. Davies, J. Monro. 1977. Association of Mg, Ca, P, and K
with various fractions in the diet, digesta, and faeces of sheep fed fresh
pasture. N.Z. J. Agric. Res. 20:441-448.
43.
Gong, J., C.W. Forsberg. 1989. Factors affecting adhesion of Fibrobacter
succinogenes S85 and adherence defective mutants to cellulose. Appl.
Environ. Microbiol. 55:3039-3044.
44.
Goetsch, A.L.; F.N. Owens. 1985. Effevcts of calcium source and level on
site of digestion and calcium levels in the digestive tract of cattle fed highconcentrate diets. J. Anim. Sci. 61:995-1003.
45.
Harfoot, C.G. 1981. Lipid metabolism in the rumen. IN: Lipid metabolism
in ruminant animals. Ed. Christie, H.H., Pergamon Press, UK. pp 21-55.
153
46.
Hartfoot, C.G., M.L. Crouchman, R.C. Noble; J.H. Moore. 1974.
Competition between fod particles and rumen bacteria in the uptake of
long-chain fatty acids and triglycerides. J. Appl. Bacteriol. 37:633.
47.
Herbaud, M-L., A. Guiseppi, F. Denizot, J. Haiech, M-C. Kilhoffer. 1998.
Calcium signaling in Bacillus subtilis. Biochim. Biophys. Acta 1448:212226.
48.
Hiltner, P., and B.A. Dehority. 1983. Effect of soluble carbohydrates on
digestion of cellulose by pure cultures of rumen bacteria. Appl. Environ.
Microbiol. 46:642-648.
49.
Hodgman, C.D., R.C. Weast, and S.M. Selby. 1959. Handbook of
chemistry and physics. Fortieth Ed. Chemical Rubber Publishing.
Cleveland, Ohio, USA. Pp 1689-1705.
50.
Hungate, R.E., and R.J. Stack. 1982. Phenylpropanoic acid: Growth factor
for Ruminococcus albus. Appl. Environ. Microbiol. 44:79-83.
51.
Hubbert, F., E. Cheng, W. Burroughs. 1958a. Mineral requirement of
rumen microorganisms for cellulose digestion in vitro. J. Anim. Sci. 17:
559-568.
52.
Hubbert, F., E. Cheng, W. Burroughs. 1958b. The influence of potassium,
sodium, rubidium, lithium and cesium on in vitro cellulose digestion by
rumen microorganisms with observations upon sodium and potassium
influences in lamb fattening rations. J. Anim. Sci. 17:576-585.
53.
Hungate, R.E. 1966. The rumen and its microbes. Academic Press. USA.
533 pp.
54.
Jasper, P.; S. Silver. 1977. Chapter 1. Magnesium transport in
microorganisms.
IN: Microrganisms and minerals. Edited by E.D.
Weinberg, Microbiology Series Vol. 3. Marcel Dekker, Inc. New York,
USA. 4-47 pp.
154
55.
Jenkins, T.C.; D.L. Palmquist. 1982. Effect of added fat and calcium on in
vitro formation of insoluble fatty acid soaps and cell wall digestibility. J.
Anim. Sci. 55:957-963.
56.
Jones, H.E., I.B. Holland, H.L. Baker, A.K. Campbell. 1999. Slow
changes in cytosolic free Ca2+ in Escherichia coli highlight two putative
influx mechanisms in response to changes in extracellular calcium. Cell
Calcium. 25: 265-274.
57.
Jones, H.E., I.B. Holland, A. K. Campbell. 2002. Direct measurement of
free Ca2+ shows different regulation of Ca2+ between the periplasm and
the cytosol of Escherichia coli. Cell Calcium. 32: 183-192.
58.
Karita, S., K. Sakka, K. Ohmiya. 1997. Cellulosomes, cellulase
complexes, of anaerobic microbes: their structure models and functions.
In: Rumen Microbes and digestive physiology in ruminants. Eds: Onodera,
R., Itabashi, H., Ushida, K., Yano, H., Saaki, Y. Japan Socity Press,
Japan. 47- 57 p.
59.
Kirby, J., J. Martin, A. Daniel, H. Flint. 1997. Dockerin-like sequences in
cellulases and xylanases from the rumen cellulolytic bacterium
Ruminococcus flavefaciens. FEMS Microbiol. Letters 149:213-219.
60.
Lamed, R., E. Setter, E.A. Bayer. 1983. Characterization of a cellulosebinding, cellulase-containing complex in Clostridium thermocellum. J.
Bacteriol. 156: 828-836.
61.
Lamed, E., R. Kenig, E. Setter. 1985. Major characteristics of the
cellulolytic system of Clostridium thermocellum coincide with those of the
purified cellulosome. Enzyme Microb. Technol. 7: 37-41.
62.
Lopez-Guisa, J.M. 1985. Corn crop residues for growing dairy heifers.
PhD Thesis, University of Wisconsin, Madison, USA. 185 pp.
63.
Luck, L.A., J.J. Falke. 1991. 19F NMR studies of the D-galactose
chemosensory receptor. 2. Ca(II) binding yields a local structural change.
Biochemistry. 30: 4257-4261.
155
64.
Lytle, B., B. Volkman, W. Westler, D. Wu. 2000. Secondary structure and
calcium-induced folding of the clostridium thermocellum dockerin domain
determined by NMR spectroscopy. Archives of Biochemistry and
Biophysics 379:237-244.
65.
Lytle, B., B. Volkman, W. Westler, M.P. Heckman, D Wu. 2001. Solution
structure of a type I dockerin domain, a novel prokaryotic, extracellular
calcium-binding domain. J. Mol. Biol. 307:745-753.
66.
Mackie, R.I., J.J. Therion. 1984. Influence of mineral interactions on
growth efficiency of rumen bacteria. IN: Herbivore Nutrition in the
Subtropics and Tropics. Ed by Gilchrist, F.M.C. and R.I. Mackie. The
Science Press. South Africa. 455-477.
67.
Maczulak, A.E.; B.A. Dehority; D.L. Palmquist. 1981. Effects of long-chain
fatty acids on growth of rumen bacteria. Appl. Envinron. Microbiol. 42:856862.
68.
Malburg, S.R.C., L.M. Malburg Jr., T. Liu, A.H. Iyo, C.W. Forsberg. 1997.
Catalytic properties of the cellulose-binding endoglucanase F from
Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 63:2449-2453.
69.
Martin, J.E., L.R. Arrington, J.E. Moore, C.B. Ammerman, G.K. Davis, R.L
Shirley. 1964. Effect of Magnesium and Sulfur upon cellulose digestion of
purified rations by cattle and sheep. J. Nutr. 33:60-64.
70.
Mayer, F. 1999. Cellular and subcellular organization of prokaryotes. IN:
Biology of the prokaryotes. Ed. by Lengeler, J.W., G. Drews, H.G.
Schegel. Thieme Stuttgart and Blackwell Science, New York. 20-46 pp.
71.
McBurney, M.I., M.S. Allen, P.J. Van Soest. 1986. Praseodymium and
copper cation-exchange capacities of neutral-detergent fibres relative to
composition and fermentation kinetics. J. Sci. Food Agric. 37:666-672.
72.
McGavin, M. C.W. Forsberg. 1988. Isolation and Characterization of
endoglucanases 1 and 2 from Bacteroides succinogenes S85. J. Bacteriol.
170:2914-2922.
156
73.
McGavin, M., J. Lam, C.W. Forsberg. 1990. Regulation and distribution of
Fibrobacter succinogenes subsp. succinogenes S85 endoglucanases.
Appl. Environ. Microbiol. 56:1235-1244.
74.
Michiels, J.M., Ch. Xi, J. Verhaert, J. Vanderleyden. 2002. The functions
of Ca+2 in bacteria: a role for E-F hand proteins? Trends Microbiol. 10:8793.
75.
Miron, J., D. Ben-Ghedalia, M. Morrison. 2001. Invited Review: Adhesion
mechanisms of rumen cellulolytic bacteria. J. Dairy Sci. 84:1294-1309.
76.
Mitsumori, M., H. Minato. 1997. Cellulose-binding proteins from rumen
microorganisms. In: Rumen Microbes and digestive physiology in
ruminants. Eds: Onodera, R., Itabashi, H., Ushida, K., Yano, H., Saaki, Y.
Japan Society Press, Japan. 35-45 p.
77.
Mitsumori, M., H. Minato. 2000. Identification of the cellulose-binding
domain of Fibrobacter succinogenes endoglucanase F. FEMS Microbiol.
Letters 183:99-103.
78.
Mitsumori, M., L.M. Xu, H. Kajikawa, M. Kurihara. 2002. Properties of
cellulose-binding modules in endoglucanase F from Fibrobacter
succinogenes S85 by means of surface plasmon resonance. FEMS
Microbiol. Letters. 214:277-281.
79.
Molloy, L.F., E.L. Richards. 1971. Complexing of calcium and magnesium
by the organic constituents of Yorkshire fog (Holcus lanatus) II.Complexing of Ca2+ and Mg2+ by cell wall fractions and organics acids. J.
Sci. Food Agric. 22: 397-402.
80.
Moncrief, M.B., M.E. Maguire. 1999. Magnesium transport in prokaryotes.
JBIC. 4:523-527.
81.
Morrison, M., J. Miron. 2000. Adhesion to cellulose by Ruminococcus
albus: a combination of cellulosomes and Pil-proteins?. FEMS Microbiol.
Letters 185:109-115.
157
82.
Nel, J.W., R.J. Moir. 1974. The effect of ruminal and duodenal application
of different levels of calcium and phosphorus to sheep on semi-purified
diets. S. Afr. J. Anim. Sci. 4:1-20.
83.
Newbold, C.J., R.J. Wallace, N.D. Watt, A.J. Richardson. 1988. Effect of
the novel ionophore tetronasin (ICI 139603) on ruminal microorganisms.
Appl. Environ. Microbiol. 54:544-547.
84.
Norris, V., S. Grant, P. Freestone, J. Canvin, F.N. Sheikh, I. Toth, M.
Trinei, K. Modha, R.I. Norman. 1996. Calcium signalling in bacteria. J.
Bacteriol. 178:3677-3682.
85.
Norris, V., G. Turnock, D. Sigee. 1996. MicroReview. The Escherichia coli
enzoskeleton. Mol. Microbiol. 19:197-204.
86.
Noziere, P. and B. Michelet-Doreau. 2000. Chapter 11. In sacco Methods.
IN: Farm Animal Metabolism and Nutrition. Ed. J.P.F. D’Mello. CAB
International Publishing. Reading, UK. 233-253 pp.
87.
Onoda, T., J. Enokizono, H. Kaya, A. Oshima, P. Freestone, V. Norris.
2000. Effects of calcium and calcium chelators on growth and morphology
of Escherichia coli L-form NC-7. J. Bacteriol. 182:1419-1422.
88.
Ordal, G.W. 1977. Calcium ion regulates chemotactic behaviour in
bacteria. Nature. 270: 66-67.
89.
Ohara H., S. Karita, T. Kimura, K. Sakka, K. Ohmiya. 2000.
Characterization of the cellulolytic complex (cellulosome) from
Ruminococcus albus. Biosci. Biotechnol. Biochem., 64:254-260.
90.
Pages S., A. Belaich, J.P. Belaich, E. Morag. R. Lamed. 1997. Speciesspecificity of the cohesin-dockerin interaction between Clostridium
thermocelum and Clostridium cellulolyticum: prediction of specificity
determinants of the dockerin domain. Proteins. 29:517-527.
158
91.
Palmquist, D.L., H.R Conrad. 1980. High fat rations for dairy cows. Tallow
and hydrolyzated blended fat at two intakes. J. Dairy Sci. 63:391-395.
92.
Palmquist, D., T. Jenkins, A.E. Joyner. 1986. Effect of dietary fat and
calcium source on insoluble soap formation in the rumen. J. Dairy Sci.
69:1020-1025.
93.
Pedgen, R., M. Larson, R. Grant, M. Morrison. 1998. Adherence of the
gram-positive bacterium Ruminococcus albus to cellulose and
identification of a novel form of cellulose-binding protein wich belongs to
the Pil family of proteins. J. Bacteriol. 180:5921-5927.
94.
Playne, M.J., M.G. Echeverria, R.G. Megarrity. 1978. Release of nitrogen,
sulphur, phosphorus, calcium, magnesium, potassium and sodium from
four tropical hays during their digestion in nylon bags in the rumen. J. Sci.
Food. Agric. 29:520-526.
95.
Poutiainen, E.. 1970. Major mineral elements in the bovine rumen fluid. I.
Concentrations and their changes between feedings. Ann. Agric. Fenniae.
9:347-356.
96.
Price, NC, and RA Dwek. 1984. Principles and problems in physical
chemistry for biochemists. Oxford University Press, 2nd edition, Hon Kong,
258 pp
97.
Rakotoarivonina, H., G. Jubelin, M. Hebraud, B. Gaillard-Martinie, E.
Forano, P. Mosoni. 2002. Adhesion to cellulose of the gram-positive
bacterium Ruminococcus albus involves type IV pili. Microbiology
148:1871-1880.
98.
Rampersaud, A., R. Utsumi, J. Delgado, S. A. Forst, M. Inouye. 1991.
Ca+2-enhanced phosphorylation of a chimeric protein kinase involved with
bacterial signal transduction. J. Biol. Chem. 266:7633-7637.
159
99.
Reusch, R.N. 2000. Review: Transmembranes ion
polyphosphate/ply-(R)-3-hydroxybutyrate
complexes.
(Mosc.) 65:280-295.
100.
Revenau, C. 2001. The effects from phenyacetic and phenyl acids on
polysaccharide degradation and growth of Ruminococcus albus. Memoire
de fin d’Etudes Ingenieur de Techniques Agricoles. Etablissement
National d’Enseignement Superior Agronomique de Dijon, France / The
Ohio State University, Ohio, USA. 50pp.
101.
Rincon, M.T., S. McRAe, J. Kirby, K. Scott, H. Flint. 2001. EndB, a
multidomain family 44 cellulase from Ruminococcus flavefaciens 17, binds
to cellulose via a novel cellulose-binding module and another R.
flavefaciens a protein via dockerin domain. Appl. Environ. Microbiol.
67:4426-4431.
102.
Rincon, M.T., S.Y. Ding, S.I. McRae, J.C. Martin, V. Aurilia, R. Lamed, Y.
Shoham, E.A. Bayer, H.J. Flint. 2003. Novel organization and divergent
dockerin specificities in the cellulosome system of Ruminococcus
flavefaciens. J. Bacteriol. 185:703-713.
103.
Roger, V., Fonty, G., Komisarczuk-Bondy, S., Gouet, P. 1990. Effects of
physiochemical factors on the adhesion to cellulose avicel of the ruminal
bacteria Ruminococcus flavefaciens and Fibrobacter succinogenes. Appl.
Environ. Microbiol. 56:3081-3087.
104.
Russell, J.B., H.J. Strobel. 1987. Effect of ionophores on ruminal
fermentation. Minireview. Appl. Environ. Microbiol. 55: 1-6.
105.
SAS / STAT®. 1999. Users’s Guide Version 8, Cary, NC; SAS Institute
Inc. (OnlineDocTM Version8,
http://www.math.wpi.edu/saspdf/stat/pdfidx.htm).
106.
Scott, H. W., and B.A. Dehority. 1965. Vitamin requirements of several
cellulolytic rumen bacteria. J. Bacteriol. 89:1169-1175.
160
transport by
Biochemistry
107.
Shapiro, L. 1993. Protein localization and asymmetry in the bacterial cell.
Cell 73:841-855.
108.
Shoham, Y., R. Lamed, E. Bayer. 1999. The cellulosome concept as an
efficient microbial strategy for the degradation of insoluble
polysaccharides. Trends Microbiol. 7:275-281.
109.
Silver, S. 1977. Chapter 2. Calcium transport in microorganisms. IN:
Microrganisms and minerals. Edited by E.D. Weinberg, Microbiology
Series Vol. 3. Marcel Dekker, Inc. New York, USA. 49-103.
110.
Sleytr, U.W., T.J. Beveridge. 1999. Bacterial S-layers. Trends Microbiol. 7:
253-260.
111.
Smith, R.J. 1995. Calcium and bacteria. Advances in Microbial Physiology
37:85-133.
112.
Smith, R.L., M.E. Maguire. 1998. Microbial magnesium transport: unusual
transporters searching for identity. Mol. Microbiol. 28:217-226.
113.
Snijder, H.J., B.W. Dijkstra. 2000. Bacterial phospholipase A: structure
and function of an integral membrane phospholipase. Biochimica et
Biophysica Acta 1488:91-101.
114.
Somers, F. 1973. The affinity of onion cell walls for calcium ions. Amer. J.
Bot. 60:987-990.
115.
Stack, R.J., R.E Hungate, and W.P Opsahl. 1983. Phenylacetic acid
stimulation of cellulose digestion by Ruminococcus albus 8. Appl.
Environ. Microbiol. 46:539-544.
116.
Sudom, A., R. Walters, L. Pastushok, D. Goldie, L. Prasad, L.T.J.
Delbaere, H. Goldie. 2003.
Mechanisms of activation of
phosphoenolpyruvate carboxykinase from Escherichia coli by Ca+2 and
desensitization by trypsin. J. Bacteriol. 185: 4233-4242.
161
117.
Sukhija, P.S., D.L Palmquist. 1990. Dissociation of calcium soaps of longchain fatty acids in rumen fluid. J. Dairy Sci. 73:1784-1787.
118.
Taylor, K.A., B. Crosby, M. McGavin, C.W. Forsberg, D.Y. Thomas. !987.
Characteristics of the endoglucanase encoded by a cel gene from
Bacteroides succinogenes expressed in Escherichia coli. Appl. Envinron.
Microbiol. 53:41-46.
119.
ThermoOrion. 2001. Calcium electrode. Instruction manual. Pp 70.
120.
Thomson, D.J., D.E Beever, M.J. Latham, M.E. Sharpe, R.A. Terry. 1978.
The effect of inclusion of mineral salts in the diet on dilution rate, the
pattern of rumen fermentation and the composition of the rumen
microflora. J. Agric. Sci., Cambridge. 91:1-7.
121.
Tisa L.S., J. Adler. 1995. Cytoplasmic free-Ca+2 level rises with repellents
and falls with attractants in Escherichia coli chemotaxis. Proc. Natl. Acad.
Sci. USA. 92:10777-10781.
122.
Torre, M., A.R. Rodriguez, F. Saura-Calixto. 1992. Study of the
interactions of calcium ions with lignin, cellulose, and pectin. J. Agric.
Food Chem. 40: 1762-1766.
123.
Van Asselt, E.J.; B.W. Dijkstra. 1999. Binding of calcium in the EF-hand of
Escherichia coli lytic transglycosylase Slt35 is important for stability. FEBS
Letters 458:429-435.
124.
Van Soest, P. 1994. Nutritional ecology of the ruminant.2nd edition. Cornell
University Press. USA. 476 p.
125.
Vinogradov, E., E.E. Egbosimba, M.B. Perry, J.S. Lam, C.W. Forsberg.
2001. Structural analysis of the carbohydrate components of the outer
membrane of the lipopolysaccharide-lacking cellulolytic ruminal bacterium
Fibrobacter succinogenes S85. Eur. J. Biochem. 268: 3566-3576.
162
126.
Wagner, K., J. Firkins, M. Eastridge, B. Hull. 1993. Replacement of corn
silage with wheat middlings and calcium chloride or sodium bicarbonate
for lactating dairy cows. J. Dairy Sci. 76:564-574.
127.
Yu, X., W. Margolin. 1997. Ca+2-mediated GTP-dependent dynamic
assembly of bacterial cell division protein FtsZ into asters and polymer
networks in vitro. EMBO J. 16: 5455-5463.
128.
Zwietering, M.H., I. Jongenburger, F.M. Rombouts, and K. Van’t Riet.
1990. Modeling of bacterial growth curve. Appl. Envinron. Microbiol.
56:1875-1881.
163